Abstract:

This article analyzes the tools and methods used in the analysis of emotions from text for the purpose of managing society. It illustrates the influence of emotions on the management of people, society and businesses processes. Their importance and the changes that occurs over time in the processes of managing society. The role that social networks plays in managing society and the methods they uses. We researched different websites and software, observed their mechanisms for data collection, its analysis and use for the purpose of managing society. In general, we analyze the most common methods, the factors that affects the choice of a particular method, their influence and based on our analy-sis, give recommendations for improving the process of analyzing emotions for the purpose of managing society.

Abstract:

BACKGROUND: Electrical impedance tomography (EIT) with indicator dilution may be clinically useful to measure relative lung perfusion, but there is limited information on the performance of this technique. METHODS: Thirteen pigs (50-66 kg) were anaesthetised and mechanically ventilated. Sequential changes in ventilation were made: (i) right-lung ventilation with left-lung collapse, (ii) two-lung ventilation with optimised PEEP, (iii) two-lung ventilation with zero PEEP after saline lung lavage, (iv) two-lung ventilation with maximum PEEP (20/25 cm HO to achieve peak airway pressure 45 cm HO), and (v) two-lung ventilation under unilateral pulmonary artery occlusion. Relative lung perfusion was assessed with EIT and central venous injection of saline 3%, 5%, and 10% (10 ml) during breath holds. Relative perfusion was determined by positron emission tomography (PET) using Gallium-labelled microspheres. EIT and PET were compared in eight regions of equal ventro-dorsal height (right, left, ventral, mid-ventral, mid-dorsal, and dorsal), and directional changes in regional perfusion were determined. RESULTS: Differences between methods were relatively small (95% of values differed by less than 8.7%, 8.9%, and 9.5% for saline 10%, 5%, and 3%, respectively). Compared with PET, EIT underestimated relative perfusion in dependent, and overestimated it in non-dependent, regions. EIT and PET detected the same direction of change in relative lung perfusion in 68.9-95.9% of measurements. CONCLUSIONS: The agreement between EIT and PET for measuring and tracking changes of relative lung perfusion was satisfactory for clinical purposes. Indicator-based EIT may prove useful for measuring pulmonary perfusion at bedside.

Abstract:

Local brain hypothermia is an attractive method for providing cerebral neuroprotection for ischemic stroke patients and at the same time reducing systemic side effects of cooling. In acute ischemic stroke patients with large vessel occlusion, combination with endovascular mechanical recanalization treatment could potentially allow for an alleviation of inflammatory and apoptotic pathways in the critical phase of reperfusion. The direct cooling of arterial blood by means of an intra-carotid heat exchange catheter compatible with recanalization systems is a novel promising approach. Focusing on the concept of "cold reperfusion", we developed an energetic model to calculate the rate of temperature decrease during intra-carotid cooling in case of physiological as well as decreased perfusion. Additionally, we discussed and considered the effect and biological significance of temperature decrease on resulting brain perfusion. Our model predicted a 2 °C brain temperature decrease in 8.3, 11.8 and 26.2 min at perfusion rates of 50, 30 and 10ml100g⋅min, respectively. The systemic temperature decrease - caused by the venous blood return to the main circulation - was limited to 0.5 °C in 60 min. Our results underline the potential of catheter-assisted, intracarotid blood cooling to provide a fast and selective brain temperature decrease in the phase of vessel recanalization. This method can potentially allow for a tissue hypothermia during the restoration of the physiological flow and thus a "cold reperfusion" in the setting of mechanical recanalization.

Abstract:

Atypical atrial flutter (AFlut) is a reentrant arrhythmia which patients frequently develop after ablation for atrial fibrillation (AF). Indeed, substrate modifications during AF ablation can increase the likelihood to develop AFlut and it is clinically not feasible to reliably and sensitively test if a patient is vulnerable to AFlut. Here, we present a novel method based on personalized computational models to identify pathways along which AFlut can be sustained in an individual patient. We build a personalized model of atrial excitation propagation considering the anatomy as well as the spatial distribution of anisotropic conduction velocity and repolarization characteristics based on a combination of a priori knowledge on the population level and information derived from measurements performed in the individual patient. The fast marching scheme is employed to compute activation times for stimuli from all parts of the atria. Potential flutter pathways are then identified by tracing loops from wave front collision sites and constricting them using a geometric snake approach under consideration of the heterogeneous wavelength condition. In this way, all pathways along which AFlut can be sustained are identified. Flutter pathways can be instantiated by using an eikonal-diffusion phase extrapolation approach and a dynamic multifront fast marching simulation. In these dynamic simulations, the initial pattern eventually turns into the one driven by the dominant pathway, which is the only pathway that can be observed clinically. We assessed the sensitivity of the flutter pathway maps with respect to conduction velocity and its anisotropy. Moreover, we demonstrate the application of tailored models considering disease-specific repolarization properties (healthy, AF-remodeled, potassium channel mutations) as well as applicabiltiy on a clinical dataset. Finally, we tested how AFlut vulnerability of these substrates is modulated by exemplary antiarrhythmic drugs (amiodarone, dronedarone). Our novel method allows to assess the vulnerability of an individual patient to develop AFlut based on the personal anatomical, electrophysiological, and pharmacological characteristics. In contrast to clinical electrophysiological studies, our computational approach provides the means to identify all possible AFlut pathways and not just the currently dominant one. This allows to consider all relevant AFlut pathways when tailoring clinical ablation therapy in order to reduce the development and recurrence of AFlut.

Abstract:

Each heartbeat is initiated by cyclic spontaneous depolarization of cardiomyocytes in the sinus node forming the primary natural pacemaker. In patients with end-stage renal disease undergoing hemodialysis, it was recently shown that the heart rate drops to very low values before they suffer from sudden cardiac death with an unexplained high incidence. We hypothesize that the electrolyte changes commonly occurring in these patients affect sinus node beating rate and could be responsible for severe bradycardia. To test this hypothesis, we extended the Fabbri et al. computational model of human sinus node cells to account for the dynamic intracellular balance of ion concentrations. Using this model, we systematically tested the effect of altered extracellular potassium, calcium, and sodium concentrations. Although sodium changes had negligible (0.15 bpm/mM) and potassium changes mild effects (8 bpm/mM), calcium changes markedly affected the beating rate (46 bpm/mM ionized calcium without autonomic control). This pronounced bradycardic effect of hypocalcemia was mediated primarily by I attenuation due to reduced driving force, particularly during late depolarization. This, in turn, caused secondary reduction of calcium concentration in the intracellular compartments and subsequent attenuation of inward I and reduction of intracellular sodium. Our in silico findings are complemented and substantiated by an empirical database study comprising 22,501 pairs of blood samples and in vivo heart rate measurements in hemodialysis patients and healthy individuals. A reduction of extracellular calcium was correlated with a decrease of heartrate by 9.9 bpm/mM total serum calcium (p < 0.001) with intact autonomic control in the cross-sectional population. In conclusion, we present mechanistic in silico and empirical in vivo data supporting the so far neglected but experimentally testable and potentially important mechanism of hypocalcemia-induced bradycardia and asystole, potentially responsible for the highly increased and so far unexplained risk of sudden cardiac death in the hemodialysis patient population.

Abstract:

Chronic kidney disease (CKD) affects 13% of the worldwide population and end stage patients often receive haemodialysis treatment to control the electrolyte concentrations. The cardiovascular death rate increases by 10% - 30% in dialysis patients than in general population. To analyse possible links between electrolyte concentration variation and cardiovascular diseases, a continuous non-invasive monitoring tool enabling the estimation of potassium and calcium concentration from features of the ECG is desired. Although the ECG was shown capable of being used for this purpose, the method still needs improvement. In this study, we examine the influence of lead reduction techniques on the estimation results of serum calcium and potassium concentrations.We used simulated 12 lead ECG signals obtained using an adapted Himeno et al. model. Aiming at a precise estimation of the electrolyte concentrations, we compared the estimation based on standard ECG leads with the estimation using linearly transformed fusion signals. The transformed signals were extracted from two lead reduction techniques: principle component analysis (PCA) and maximum amplitude transformation (Max- Amp). Five features describing the electrolyte changes were calculated from the signals. To reconstruct the ionic concentrations, we applied a first and a third order polynomial regression connecting the calculated features and concentration values. Furthermore, we added 30 dB white Gaussian noise to the ECGs to imitate clinically measured signals. For the noisefree case, the smallest estimation error was achieved with a specific single lead from the standard 12 lead ECG. For example, for a first order polynomial regression, the error was 0.0003±0.0767 mmol/l (mean±standard deviation) for potassium and -0.0036±0.1710 mmol/l for calcium (Wilson lead V1). For the noisy case, the PCA signal showed the best estimation performance with an error of -0.003±0.2005 mmol/l for potassium and -0.0002±0.2040 mmol/l for calcium (both first order fit). Our results show that PCA as ECG lead reduction technique is more robust against noise than MaxAmp and standard ECG leads for ionic concentration reconstruction.

Abstract:

Changes of serum and extracellular ion concentrations occur regularly in patients with chronic kidney disease (CKD). Recently, hypocalcemia, i.e. a decrease of the extra-cellular calcium concentration [Ca2+]o, has been suggested as potential pathomechanism contributing to the unexplained high rate of sudden cardiac death (SCD) in CKD patients. In particular, there is a hypothesis that hypocalcaemia could slow down natural pacemaking in the human sinus node to fatal degrees. Here, we address the question whether there are inter-species differences in the response of cellular sinus node pacemaking to changes of [Ca2+]o. Towards this end, we employ computational models of mouse, rabbit and human sinus node cells. The Fabbri et al. human model was updated to consider changes of intracellular ion concentrations. We identified crucial inter-species differences in the response of cellular pacemaking in the sinus node to changes of [Ca2+]o with little changes of cycle length in mouse and rabbit models (<83 ms) in contrast to a pronounced bradycardic effect in the human model (up to > 1000 ms). Our results suggest that experiments with human sinus node cells are required to investigate the potential mechanism of hypocalcaemia-induced bradycardic SCD in CKD patients and small animal models are not well suited.

Abstract:

Atrial fibrillation (AF) is the most prevalent form of cardiac arrhythmia. The atrial wall thickness (AWT) can potentially improve our understanding of the mechanism underlying atrial structure that drives AF and provides important clinical information. However, most existing studies for estimating AWT rely on ruler-based measurements performed on only a few selected locations in 2D or 3D using digital calipers. Only a few studies have developed automatic approaches to estimate the AWT in the left atrium, and there are currently no methods to robustly estimate the AWT of both atrial chambers. Therefore, we have developed a computational pipeline to automatically calculate the 3D AWT across bi-atrial chambers and extensively validated our pipeline on both ex vivo and in vivo human atria data. The atrial geometry was first obtained by segmenting the atrial wall from the MRIs using a novel machine learning approach. The epicardial and endocardial surfaces were then separated using a multi-planar convex hull approach to define boundary conditions, from which, a Laplace equation was solved numerically to automatically separate bi-atrial chambers. To robustly estimate the AWT in each atrial chamber, coupled partial differential equations by coupling the Laplace solution with two surface trajectory functions were formulated and solved. Our pipeline enabled the reconstruction and visualization of the 3D AWT for bi-atrial chambers with a relative error of 8% and outperformed existing algorithms by >7%. Our approach can potentially lead to improved clinical diagnosis, patient stratification, and clinical guidance during ablation treatment for patients with AF.

Abstract:

Under persistent atrial fibrillation (peAF), cardiac tissue experiences electrophysiological and structural remodeling. Fibrosis in the atrial tissue has an important impact on the myocyte action potential and its propagation. The objective of this work is to explore the effect of heterogeneities present in the fibrotic tissue and their impact on the intracardiac electrogram (EGM). Human atrial myocyte and fibroblast electrophysiology was simulated using mathematical models proposed by Koivumäki et al. to represent electrical remodeling under peAF and the paracrine effect of the transforming grow factor 1 (TGF-1). 2D tissue simulations were computed varying the density of fibrosis (10%, 20% and 40%), myofibroblasts and collagen were randomly distributed with different ratios (0%-100%, 50%-50% and 100%- 0%). Results show that increasing the fibrosis density changes the re-entry dynamics from functional to anatomical due to a block in conduction in regions with high fibrosis density (40%). EGM morphology was affected by different ratios of myofibroblasts-collagen. For low myofibroblast densities (below 50%) the duration of active segments was shorter compared to higher myofibroblasts densities (above 50%). Our results show that fibrosis heterogeneities can alter the dynamics of the re-entry and the morphology of the EGM.

Abstract:

Fluorescence video angiography is used in neurosurgery to intraoperatively monitor the vascular func-tion, namely the blood flow. This is done by injecting the dye Indocyanine green (ICG) intravenously. After excitation by a near-infrared light source, the fluorescence signal is captured by a camera system. The recorded signal is used to qualitatively assess the vascular function during the intervention. This provides the surgeon with an immediate feedback of the quality of his surgery. Nevertheless, this qualitative assessment needs to be extended and a quantitative value should be calculated to assist the surgical staff. This step requires a standardized and validated test setup mimicking cerebral vessels for studies, such as measurement of the flow and flow profile. This includes the confirmation of the suita-bility of the investigation site in the phantom. Therefore, a flow phantom is designed according to the requirements and set up. The requirements include a variable diameter of the vessel mimicking tubes, variable flow range within the clinical relevant range, a handy and precise injection system with an ini-tial ICG concentration which minimizes quenching effects, a non-toxic and low cost blood analogue with similar viscosity as human blood and finally a last requirement which need more explanation. Re-al blood should not be used due to the contamination of the pump, so water is used as flow media. But the ICG is dissolved in a protein solution and should be surrounded by a protein solution to ensure mixing and diffusion into the same solution media, so the ICG should not get into touch with the flow media water. The investigation sites are given in the ranges which are confirmed to be suitable. The flow phantom provides a consistent testing environment and will be used to conduct studies analyzing the suitability of different methods to assess the flow by fluorescence imaging.

Abstract:

We suggest a new regularization method for reconstruction of cardiac transmembrane voltages (TMV) from body surface potentials that is based on imposing similarity between time-aligned TMVs. An iterative scheme is proposed to update the delays needed for time-alignment. Evaluation of the method using simulated ventricular pacings showed a clear improvement over second order Tikhonov.

Abstract:

Activation times (AT) describe the sequence of cardiac depolarization and represent one of the most important parameters for analysis of cardiac electrical activity. However, estimation of ATs can be challenging due to multiple sources of noise such as fractionation or baseline wander. If ATs are estimated from signals reconstructed using electrocardiographic imaging (ECGI), additional problems can arise from over-smoothing or due to ambiguities in the inverse problem. Often, resulting AT maps show falsely homogeneous regions or artificial lines of block. As ATs are not only important clinically, but are also commonly used for evaluation of ECGI methods, it is important to understand where these errors come from. We present results from a community effort to compare methods for AT estimation on a common dataset of simulated ventricular pacings. ECGI reconstructions were performed using three different surface source models: transmembrane voltages, epi-endo potentials and pericardial potentials, all using 2nd-order Tikhonov and 6 different regularization parameters. ATs were then estimated by the community participants and compared to the ground truth. While the pacing site had the largest effect on AT correlation coefficients (CC larger for lateral than for septal pacings), there were also differences between methods and source models that were poorly reflected in CCs. Results indicate that artificial lines of block are most severe for purely temporal methods. Compared to the other source models, ATs estimated from transmembrane voltages are more precise and less prone to artifacts.

Abstract:

ECG imaging estimates the cardiac electrical activity from body surface potentials. As this involves solving a severly ill-posed problem, additional information is required to get a unique and stable solution. Recent progress is based on introducing more problem-specific information by exploiting the structure of cardiac excitation. However, added information must be either certain or general enough to not impair the solution. We have recently developed a method that uses a spatio-temporal basis to restrict the solution space. In the present work, we analyzed this method with respect to one of the most fundamental assumptions made during basis creation: cardiac (an)isotropy. We tested the reconstruction using simulations of ventricular pacings and then applied it to clinical data. In simulations, the overall median localization error was smallest with a basis including fiber orientation. For the clinical data, however, the overall error was smallest with an isotropic basis. This observation suggests that modeling priors should be introduced with care, whereby further work is needed.

Abstract:

The SuLMaSS project [1] will advance, develop, build, evaluate, and test infrastructure for sustainable lifecycle management of scientific software. The infrastructure is tested and evaluated by an existing cardiac electrophysiology simulation software project, which is currently in the prototype state and will be advanced towards optimal usability and a large and active user community. Thus, SuLMaSS is focused on designing and implementing application-oriented e-research technologies and the impact is three-fold: - Provision of a high quality, user-friendly cardiac electrophysiology simulation software package that accommodates attestable needs of the scientific community. - Delivery of infrastructure components for testing, safe-keeping, referencing, and versioning of all phases of the lifecycle of scientific software. - Serve as a best practice example for sustainable scientific software management. Scientific software development in Germany and beyond shall benefit through both the aforementioned best practice role model and the advanced infrastructure that will, in part, be available for external projects as well. With adding value for the wider scientific cardiac electrophysiology community, the software will be available under an open source license and be provided for a large share of people and research groups that can potentially leverage computational cardiac modeling methods. Institutional infrastructure will be extended to explore, evaluate and establish the basis for research software development regarding testing, usage, maintenance and support. The cardiac electrophysiology simulator will drive and showcase the infrastructure formation, thus serving as a lighthouse project. The developed infrastructure can be used by other scientific software projects in future and aims to support the full research lifecycle from exploration through conclusive analysis and publication, to archival, and sharing of data and source code, thus increasing the quality of research results. Moreover it will foster a community-based collaborative development and improve sustainability of research software.

Abstract:

Electrical impedance tomography is an accepted and validated tool to analyze and support mechanical ventilation at the bedside. In the future it could furthermore clinically provide information of the pulmonary perfusion and other blood volume changes within the thorax by exploiting a cardiosynchronous EIT component. In the presented study, the spatial forward sensitivity against different background lung tissue distributions was analyzed. Spheres with a 10% change of the background conductivity were introduced in the lungs and in the heart. The cranio-caudal distribution of sensitivity had a bell shape and was similar between all simulated scenarios, varying only in magnitude. If the background tissue conductivity within the lungs was chosen to be the one of deflated tissue, the overall sensitivity was 46% smaller compared to the overall sensitivity against inflated lung tissue conductivity. Within the heart region, the sensitivity was increased for fully deflated lung tissue conductivity (23% relative to the sensitivity in the lungs) compared to a homogeneous distribution of inflated lung tissue conductivity (10% relative to the sensitivity in the lungs).

Abstract:

The heart acts as the pump of the cardiovascular system due to the active stress developed in individ- ual cardiac muscle cells. The spatio-temporal distribution of this active stress could contain relevant diagnostic information but can currently not be measured in vivo. We introduce a method to esti- mate dynamic cardiac active stress fields from endocardial surface motion tracking derived from e.g. magnetic resonance imaging data. This ill-posed non-linear problem is solved using Tikhonov regu- larization in space and time in conjunction with a continuum mechanics forward model. We present a proof-of-concept using data from a biophysically detailed multiscale model of cardiac electrome- chanics (7649 tetrahedral elements) in which we could accurately reproduce cardiac motion (surface error <0.4 mm) and identify non-contracting regions due to myocardial infarction scars (active stress error <10 kPa). This inverse method could eventually be used to non-invasively derive personalized diagnostic information in terms of dynamic active stress fields which are not accessible today.

Abstract:

Electrocardiographic Imaging (ECGI) requires robust ECG forward simulations to accurately calculate cardiac activity. However, many questions remain regarding ECG forward simulations, for instance: there are not common guidelines for the required cardiac source sampling. In this study we test equivalent double layer (EDL) forward simulations with differing cardiac source resolutions and different spatial interpolation techniques. The goal is to reduce error caused by undersampling of cardiac sources and provide guidelines to reduce said source undersampling in ECG forward simulations. Using a simulated dataset sampled at 5 spatial resolutions, we computed body surface potentials using an EDL forward simulation pipeline. We tested two spatial interpolation methods to reduce error due to undersampling triangle weighting and triangle splitting. This forward modeling pipeline showed high frequency artifacts in the predicted ECG time signals when the cardiac source resolution was too low. These low resolutions could also cause shifts in extrema location on the body surface maps. However, these errors in predicted potentials can be mitigated by using a spatial interpolation method. Using spatial interpolation can reduce the number of nodes required for accurate body surface potentials from 9,218 to 2,306. Spatial interpolation in this forward model could also help improve accuracy and reduce computational cost in subsequent ECGI applications.

Abstract:

The boundary element method is widely used to solve the forward problem of electrocardiography, i.e. to calculate the body surface potentials (BSP) caused by the heart’s electrical activity. This requires discretization of boundary surfaces between compartments of a torso model. Often, the resolution of the surface bounding the heart is chosen above 1 mm, which can lead to spikes in resulting BSPs. We demonstrate that this artifact is caused by discontinuous propagation of the wavefront on coarse meshes and can be avoided by blurring cardiac sources before spatial downsampling. We evaluate different blurring methods and show that Laplacian blurring reduces the BSP error 5-fold for both transmembrane voltages and extracellular potentials downsampled to 3 different resolutions. We suggest a method to find the optimal blurring parameter without having to compute BSPs using a fine mesh.

Abstract:

Heterogeneous atrial substrate can induce, maintain and promote cardiac arrhythmias. The level of heterogeneity may be used to assess disease progression. One key parameter, suspected to be correlated with tissue vitality is the conduction velocity (CV). By measuring not only the current CV of the patient but rather its rate dependent changes, restitution information is gained. In the following, we show our approach towards a patient-specific quantitative atrial substrate characterization by combining sets of local and global CV restitution measurements to create a parametrization of the individual patient substrate characteristics.

Abstract:

Stroke is the third-most cause of death in developed countries. A new promising treatment method in case of an ischemic stroke is selective intracarotid blood cooling combined with mechanical artery recanalization. However, the control of the treatment requires invasive or MRI-assisted measurement of cerebral temperature. An auspicious alternative is the use of computational modeling. In this work, we extended an existing 1D hemodynamics model including the characteristics of the anterior, middle and posterior cerebral artery. Furthermore, seven ipsilateral anastomoses were additionally integrated for each hemisphere. A potential stenosis was placed into the M1 segment of the middle cerebral artery, due to the highest risk of occlusion there. The extended model was evaluated for various degrees of collateralization (“poor”, “partial” and “good”) and degrees of stenosis (0%, 50%, 75% and 99.9%). Moreover, cerebral autoregulation was considered in the model. The higher the degree of collateralization and the degree of stenosis, the higher was the blood flow through the collaterals. Hence, a patient with a good collateralization could compensate a higher degree of occlusion and potentially has a better outcome after an ischemic stroke. For a 99.9% stenosis, an increased summed mean blood flow through the collaterals of +97.7% was predicted in case of good collateralization. Consequently, the blood supply via the terminal branches of the middle cerebral artery could be compensated up to 44.4% to the physiological blood flow. In combination with a temperature model, our model of the cerebral collateral circulation can be used for tailored temperature prediction for patients to be treated with selective therapeutic hypothermia.

Abstract:

In Western countries, stroke is the third-most cause of death; 35- 55% of the survivors experience permanent disability. Mild therapeutic hypothermia (TH) showed neuroprotective effect in patients returning from cardiac arrest and is therefore assumed to decrease stroke induced cerebral damage. Recently, an intracarotid cooling sheath was developed to induce local TH in the penumbra using the cooling effect of cerebral blood flow via collaterals. Computational modeling provides unique opportunities to predict the resulting cerebral temperature without invasive procedures. In this work, we generated a simplified brain model to establish a cerebral temperature calculation using Pennes’ bio-heat equation and a 1D hemodynamics model of the cranial artery tree. In this context, we performed an extensive literature research to assign the terminal segments of the latter to the corresponding perfused tissue. Using the intracarotid cooling method, we simulated the treatment with TH for different degrees of stenosis in the middle cerebral artery (MCA) and analyzed the resulting temperature spatialtemporal distributions of the brain and the systemic body considering the influence of the collaterals on the effect of cooling.

Abstract:

Quantifying the atrial conduction velocity (CV) reveals important information for targeting critical arrhythmia sites that initiate and sustain abnormal electrical pathways, e.g. during atrial flutter. The knowledge about the local CV distribution on the atrial surface thus enhances clinical catheter ablation procedures by localizing pathological propagation paths to be eliminated during the intervention. Several algorithms have been proposed for estimating the CV. All of them are solely based on the local activation times calculated from electroanatomical mapping data. They deliver false values for the CV if applied to regions near scars or wave collisions. We propose an extension to all approaches by including a distinct preprocessing step. Thereby, we first identify scars and wave front collisions and provide this information for the CV estimation algorithm. In addition, we provide reliable CV values even in the presence of noise. We compared the performance of the Triangulation, the Polynomial Fit and the Radial Basis Functions approach with and without the inclusion of the aforementioned preprocessing step. The evaluation was based on different activation patterns simulated on a 2D synthetic triangular mesh with different levels of noise added. The results of this study demonstrate that the accuracy of the estimated CV does improve when knowledge about the depolarization pattern is included. Over all investigated test cases, the reduction of the mean velocity error quantified to at least 25 mm/s for the Radial Basis Functions, 14 mm/s for the Polynomial Fit and 14 mm/s for the Triangulation approach compared to their respective implementations without the preprocessing step. Given the present results, this novel approach can contribute to a more accurate and reliable CV estimation in a clinical setting and thus improve the success of radio-frequency ablation to treat cardiac arrhythmias.

Abstract:

Optical image processing is part of many applications used for brain surgeries. Microscope camera, or patient movement, like brain-movement through the pulse or a change in the liquor, can cause the image processing to fail. One option to compensate movement is feature detection and spatial allocation. This allocation is based on image features. The frame wise matched features are used to calculate the transformation matrix. The goal of this project was to evaluate different feature detectors based on spatial density and temporal robustness to reveal the most appropriate feature. The feature detectors included corner-, and blob-detectors and were applied on nine videos. These videos were taken during brain surgery with surgical microscopes and include the RGB channels. The evaluation showed that each detector detected up to 10 features for nine frames. The feature detector KAZE resulted in being the best feature detector in both density and robustness.

Abstract:

Current surgical microscope systems have excellent optical properties but still involve some limitations. A future fully digital surgical microscope may overcome some major limitations of typical optomechanical systems, like ergonomic restrictions or limited number of observers. Furthermore, it can leverage and provide the full potential of digital reality. To achieve this, the frontend, the reconstruction of the digital twin of the surgical scenery, as well as the backend, the 3-D visualization interface for the surgeon, need to work in real-time. To investigate the visualization chain, we developed a virtual reality environment allowing pretesting this new form of 3-D data presentation. In this study, we wanted to answer the following question: How must the visualization pipeline look like to achieve a real-time update of the 3-D digital reality scenery. With our current approach, we were able to obtain visualizations with a frame rate of 120 frames per second and a 3-D data update rate of approximately 90 datasets per second. In a further step, a first prototype of a real-time mixed-reality head mounted visualization system could be manufactured based on the knowledge gained during the virtual reality pretesting.

Abstract:

Today the vascular function after interventions as Bypass surgeries are checked qualitatively by observing the blood dynamics inside the vessel via Indocyanine Green (ICG) Fluorescence Angiography. This state-of-the-art should be upgraded and has to be improved and converted towards a quantitatively measured blood flow. Previous approaches show that the blood flow measured from fluorescence angiography cannot be easily calibrated to a gold standard reference. In order to systematically address the possible source of error we investigate as a first step the discretization error in a camera-based measurement of the vessel’s geometry. In order to generate an error-free ground truth, a vessel model has been developed based on mathematical functions. This database is then used to determine the error in discretizing the centerline of the structure and estimate its effects on the accuracy of the flow calculation. As result the model is implemented according to the conditions which are set up to ensure transferability on camera-based segmentations of vessels. In this paper the relative discretization error for estimating the centerline length of segmented vessels could be calculated in the range of 6.3%. This would reveal significant error propagated to the estimation of the blood flow value derived by camera-based angiography.

Abstract:

Changes of serum and extracellular ion concentrations occur regularly in patients with chronic kidney disease (CKD). Recently, hypocalcaemia, i.e. a decrease of the extracellular calcium concentration [Ca 2+ ] o , has been suggested as potential pathomechanism contributing to the unexplained high rate of sudden cardiac death (SCD) in CKD patients. In particular, there is a hypothesis that hypocalcaemia could slow down natural pacemaking in the human sinus node to fatal degrees. Here, we address the question whether there are inter-species differences in the response of cellular sinus node pacemaking to changes of [Ca 2+ ] o . Towards this end, we employ computational models of mouse, rabbit and human sinus node cells. The Fabbri et al. human model was updated to consider changes of intracellular ion concentrations. We identified crucial inter-species differences in the response of cellular pacemaking in the sinus node to changes of [Ca 2+ ] o with little changes of cycle length in mouse and rabbit models (<83 ms) in contrast to a pronounced bradycardic effect in the human model (up to >1000 ms). Our results suggest that experiments with human sinus node cells are required to investigate the potential mechanism of hypocalcaemia-induced bradycardic SCD in CKD patients and small animal models are not well suited.

Abstract:

In Western countries, stroke is the third-most widespread cause of death. 80% of all strokes are ischemic and show a mortality rate of about 25%. Furthermore, 35-55% of affected patients retain a permanent disability. Therapeutic hypothermia (TH) could decrease inflammatory processes and the stroke-induced cerebral damage. Currently, the standard technique to induce TH is cooling of the whole body, which can cause several side effects. A novel cooling sheath uses intra-carotid blood cooling to induce local TH. Unfortunately, the control of the temporal and spatial cerebral temperature course requires invasive temperature measurements. Computational modeling could be used to predict the resulting temperature courses instead. In this work, a detailed 1D hemodynamics model of the cerebral arterial system was coupled with an energetic temperature model. For physiological conditions, 50% and 100% M1-stenoses, the temperatures in the supply area of the middle cerebral artery (MCA) and of the systemic body was analyzed. A 2K temperature decrease was reached within 10min of cooling for physiological conditions and 50% stenosis. For 100% stenosis, a significant lower cooling effect was observed, resulting in a maximum cerebral temperature decrease of 0.7K after 30min of cooling. A significant influence of collateral flow rates on the cooling effect was observed. However, regardless of the stenosis degree, the temperature decrease was strongest within the first 20min of cooling, which demonstrates the fast and effective impact of intra-carotid blood cooling.

Abstract:

In Europe, the prevalence of chronic kidney disease lay at approximately 18.38% in 2016. A common treatment for patients in the end stage of this disease is haemodialysis. However, patients undergoing this therapy suffer from an increased risk of cardiac death. A hypothesis is that the cause is an inbalanced electrolyte concentration. To study the underlying mechanisms of this phenomenon and fight the consequences, a continous non-invasive monitoring technique is desired. In this work, we investigated the possibility to reconstruct the extracellular concentrations of potassium and calcium from ECG signals. Therefore, we extracted 71 ECGs using the simulation results of a modified Himeno et al. ventricular cell model comprising variations of the extracellular ionic concentrations of potassium and calcium. The changes dependent on the different extracellular ionic concentrations were captured with five ECG features. These were used to train an artificial neural network for regression. The study was performed both for noise-free and noisy data. The estimation error for the reconstruction of the potassium concentrations was -0.01±0.14 mmol/l (mean±standard deviation) in the noise- free case, -0.03±0.46mmol/l in the noisy case (30dB SNR). For calcium, the result was 0.01±0.11mmol/l in the noise- free case, 0.02±0.17mmol/l in the noisy case. For both ion types, the result was improved by augmenting the dataset. We therefore conclude that with the calculated features, we are able to reconstruct the extracellular ionic concentrations for both potassium and calcium with an acceptable precision. When analysing noisy signals, the accuracy of the estimation method is still sufficient but can be further improved by an augmentation of the dataset.

Abstract:

The risk of sudden cardiac death (SCD) is increased 14-fold in chronic hemodialysis (HD) patients compared to patients with normal kidney function suffering from cardiovascular diseases. This high rate is not explained by traditional cardiovascular risk factors. Recently, severe bradycardia has been implicated in SCD in HD patients. Mathematical modelling suggests an electrophysiological link between low serum calcium (Ca) levels and bradycardia. Therefore, we analyzed the correlation between heart rate (HR) and Ca as well as potassium (K).

Abstract:

The outcomes of ablation targeting either reentry activations or fractionated activity during persistent atrial fibrillation (AF) therapy remain suboptimal due to, among others, the intricate underlying AF dynamics. In the present work, we sought to investigate such AF dynamics in a heterogeneous simulation setup using recurrence quantification analysis (RQA). AF was simulated in a spherical model of the left atrium, from which 412 unipolar atrial electrograms (AEGs) were extracted (2 s duration; 5 mm spacing). The phase was calculated using the Hilbert transform, followed by the identification of points of singularity (PS). Three regions were defined according to the occurrence of PSs: 1) no rotors; 2) transient rotors and; 3) long-standing rotors. Bipolar AEGs (1114) were calculated from pairs of unipolar nodes and bandpass filtered (30-300 Hz). The CARTO criterion (Biosense Webster) was used for AEGs classification (normal vs. fractionated). RQA attributes were calculated from the filtered bipolar AEGs: determinism (DET); recurrence rate (RR); laminarity (LAM). Sample entropy (SampEn) and dominant frequency (DF) were also calculated from the AEGs. Regions with longstanding rotors have shown significantly lower RQA attributes and SampEn when compared to the other regions, suggesting a higher irregular behaviour (P≤0.01 for all cases). Normal and fractionated AEGs were found in all regions (respectively; Region 1: 387 vs. 15; Region 2: 221 vs. 13; Region 3: 415 vs. 63). Region 1 vs. Region 3 have shown significant differences in normal AEGs (P≤0.0001 for all RQA attributes and SampEn), and significant differences in fractionated AEGs for LAM, RR and SampEn (P=0.0071, P=0.0221 and P=0.0086, respectively). Our results suggest the co-existence of normal and fractionated AEGs within long-standing rotors. RQA has unveiled distinct dynamic patterns–irrespective of AEGs classification–related to regularity structures and their nonstationary behaviour in a rigorous deterministic context.

Abstract:

Intracardiac electrograms (EGMs) form the basis for the diagnosis of arrhythmia mechanisms. Bipolar EGMs dominate clinical practice despite major disadvantages over unipolar EGMs since noise strongly distorts the latter. In this study, we quantified and reduced the noise level of uni- and bipolar EGMs recorded with Rhythmia HDx and the Orion catheter. Distinct noise frequencies in the power spectral density (PSD) were detected with a sliding win- dow of variable width and subsequently removed by notch filtering. The absolute peak to peak voltage remaining in the inactive segments after baseline removal quantified the noise level of the system. An international, multi-center selection of 33 patients served as a broad sample cohort. The case-specific detection and removal of noise peaks reduced the noise level in unipolar EGMs by 30% to 0.076 mV compared to standard clinical filtering. With a bipolar noise level of 0.01 mV, we saw that Rhythmia HDx meets the low noise floor claimed in the system specifica- tions. Certain noise frequencies presented permanently in all cases whereas others showed up only intermittently or in individual cases. The suggested extension of filter settings lowers the noise level, enhances the detailed segmentation of low volt- age areas, and encourages to exploit the advantages of unipolar over bipolar EGMs in clinical practice.

Abstract:

Improved understanding of the effects of variability in electrophysiological activity within the human heart is key to understanding and predicting cardiovascular response to disease and treatments. Previous studies have considered either regional variation in action potentials or inter-subject variability within a single region of the atria. In this study, we hypothesize that the regional differences in morphology derive not only from variation in dependence on individual conductances, but also from the relationship between multiple conductances. Using the Monte-Carlo Sampling Method and the Maleckar cellular model for electrophysiology, we created an In-Silico Population of Models. Each conductance was varied +/- 100% from the standard model. The population was divided into regional groups based on biomarkers. Results showed regional variation in the dependence on relationships between conductances. In the right atrial appendage the value of gK1 was found to be only twice as influential as the relationship between gK1 and gKur on the APD90 biomarker. Other relationships that had a significant impact included gTo-gKur; gKr-gK1; gNaK- gNaCa and gKur-gNaK for various regions. R2 values for first order linear regression models showed significant relationships were left out in the analysis. This was significantly improved in the second order R2 values.

Abstract:

Several works have studied arrhythmogenicity of a given atrial model using different methods to initiate and simulate the perpetuation of re-entrant activity. We evaluated and compared two state-of-the-art methods showing their influence on the estimated vulnerability to arrhythmia of an atrial model.

Abstract:

In silico studies are often used to analyze mechanisms of cardiac arrhythmias. The electrophysiological cell models that are used to simulate the membrane potential in these studies range from highly detailed physiological models to simplistic phenomenological models. To effectively cover the middle ground between those cell models, we utilize the manifold boundary approxi- mation method (MBAM) to systematically reduce the widely used O’Hara-Rudy ventricular cell model (ORd) and investigate the influence of parametrization of the model as well as different strategies of choosing input quantities, further called quantities of interest (QoI). As a result of the reduction process, we present three re- duced model variants of the ORd model that only contain a fraction of the original model’s ionic currents resulting in a twofold speedup in computation times compared to the original model. We find that the reduced models show similar action potential duration restitution and repolarization rates. Additionally, we are able to initialize and observe stable spiral wave dynamics on a 3D tissue patch for 2 out of the 3 reduced models.

Abstract:

Catheter ablation targeting low voltage areas (LVA) is commonly being used to treat atrial fibrillation (AF) in pa- tients with persistent AF. However, it is not always certain that the areas marked as low voltage (LV) are correct. This can be related to how the voltage is calculated. There- fore, this paper focuses on comparing different calculation methods, specifically, with regards to spatial distribution. Two voltage maps obtained in AF were used, removing points which did not meet the required specifications. The peaks for the remaining points, in regions of the left atrium, were then found and the voltage was calculated based on taking the peak to peak (p2p) for different beats. For around 30% of the points on the map, the voltage only changed by 0.1mV when taking one beat versus all beats. However, for some individual points, the difference was substantial, around 0.8mV, depending on the beat cho- sen. Additionally, the inter-method variability increased by around 0.1mV when considering all methods compared to only methods calculated using more than one point. It was found that taking a method which considers all p2p values would be a more appropriate method for cal- culating the voltage. Thus, providing a technique, which could improve the accuracy of identifying LVA in an AF map.

Abstract:

Nowadays, a large share of the global population is affected by heart rhythm disorders. Computational modelling is a useful tool for understanding the dynamics of cardiac arrhythmias. Several recent clinical and experimental studies suggest that atrial fibrillation is maintained by re-entrant drivers (e.g. rotors). As a consequence, numerous works have addressed atrial arrhythmogenicity of a given electrophysiological model using different methods to simulate the perpetuation of re-entrant activity. However, no common procedure to test atrial fibrillation vulnerability has yet been defined. Here, we systematically evaluate and compare two state-of-the-art methods. The first one is rapid extrastimulus pacing from rim of the four pulmonary veins. The second consists of placing phase singularities in the atria, estimating an activation time map by solving the Eikonal equation and finally using this as initial condition for the electrical cardiac propagation simulation. In this way, we are forcing the wavefronts to follow re-entrant circuits with low computational cost thus less simulation time. We aim to identify a methodology to quantify arrhythmia vulnerability on patient-specific atrial geometries and substrates. We will proceed with in-silico experiments, comparing the results of these two methods to initiate re-entrant activity, checking the influence of the different parameters on the dynamics on the re-entrant drivers and finally extracting a valid set of parameters allowing to reliably assess re-entry vulnerability. The final objective is to come up with an easily reproducible minimal set of simulations to assess vulnerability of a particular atrial substrate (cellular and tissue model) or of distinct anatomical atrial geometries to arrhythmic episodes. Given the great need of exploring susceptibility to atrial arrhythmias, i.e. after a first ablation procedure, this study can provide a useful tool to test new treatment strategies and to learn how to prevent the onset and progression of atrial fibrillation.

Abstract:

Atrial fibrillation is the most common cardiac arrhythmia characterized by a rapid and irregular atrial excitation rate. Mimicking this behaviour, the S1-S2 stimulation protocol is currently the clinically established method for measuring tissue rate dependency, leading to a need for an automated segmentation method. We propose a method for stimulus artefact removal tailored towards the S1-S2 protocol. We show that this method results in the detection of atrial signals minimizing distortion by the stimulus artefact and is therefore an effective segmentation tool and a building block for automation of signal analysis.

Abstract:

A catheter has a mapping assembly having a plurality of splines mounted at its distal portion. The splines each have a proximal end disposed at the distal portion of the catheter body and a distal end and configured as a Fibonacci spiral arm that diverges outwardly from the proximal end. The splines have a support arm with shape memory, a non-conductive covering in surrounding relation to the support arm, at least one location sensor mounted at or near the distal end, a plurality of electrodes mounted in surrounding relation to the non-conductive covering, and a plurality of electrode lead wires extending within the non-conductive covering. Each electrode lead wire is attached to a corresponding one of the electrodes.

Abstract:

Catheter ablation is a major treatment for atrial tachycardias. Hereby, the precise monitoring of the lesion formation is an important success factor. This book presents computational, wet-lab, and clinical studies with the aim of evaluating the signal characteristics of the intracardiac electrograms (IEGMs) recorded around ablation lesions from different perspectives. The detailed analysis of the IEGMs can optimize the description of durable and complex lesions during the ablation procedure.

Abstract:

The acquisition of electroanatomical mapping data with clinically relevant information on the atrial substrate still poses a challenging problem in the field of medical and med- ical engineering research. In recent times, studies about the use of a former ablation catheter for the purpose of local impedance (LI) measurements provided promising findings. For the extraction of substrate characteristics, however, influencing factors without relevant information have to be removed from clinical LI signals. The primary aim of this work is to identify influencing factors contributing to the measured LI signals. To perform substrate mapping, it is important to know which of the influencing factors contain relevant information for substrate mapping and which factors only flaw the measurement. Lastly, this work investigates whether the identified influencing factors can be detected, quantified, and removed, and to which degree they affect the LI compared to each other. This thesis aims at identifying all influencing factors of LI measurements using signal processing of clinical data, by the analysis of in vitro measurements, and with the help of catheter simulations. The impact of factors like the catheter movement, the distance to the endocardial surface and overall orientation were shown in both signal processing and simulation. Recurring phenomena like the artifacts caused by steerable sheath segments and the distortions caused by the cardiac rhythm were analyzed. The influence of sodium chloride solution irrigation and the influence of the concentration of the solution commonly used in clinical practice were determined. Approaches to quantify the behavior of initial fast impedance decreases in both in vitro and clinical measurements were implemented in catheter simulations using spherical irrigation volumes. Regarding the influence of cardiac geometry, simulations showed significant LI differences depending on the surrounding tissue settings. The analysis of patho- logical influences showed exemplary substrate LI values of fibrotic and fatty regions which could be distinguished by LI mapping in future applications. Challenges in the current clinical system with special regard to recommendations for future hardware optimization were discussed. In conclusion, it was shown that the identification of influences on the LI signal is possible and that currently the lack of information on real time anatomic data leads to flawed mapping results. Therefore, pathological substrate is not yet distinguishable from healthy tissue with LI mapping.

Abstract:

Surgical microscopes are important instruments in medicine. Their function is the upscaling of images in aid of surgeries. The employment of more than one camera allows for 3D reconstruction of the operation site. A prerequisite, however, is the synchronization of said cameras. Furthermore, the pictures taken must be of equal quality, i.e. cameras of the same type must be used. Also, their colour spaces should be equally calibrated due to differences in production and colour temperature. For this, an understanding of the manner in which the human eye works is needed, as the cameras work in a similar way. Furthermore, the reader is introduced to the technical fundamentals of camera function, e.g. Bayer filters and automatic white balance and made familiar with the basics of software design and user interface creation with C# and Windows Forms in Visual Studio. In addition, the tools used in this work and the custom user interfaces which were created are presented. In addition to the fundamental technical background, the different factors that influence the cameras colour accuracy are discussed and ways to synchronize the cameras by using timestamps and a metronome are presented. Finally, the reasons which caused this approach to not work as expected in this case are presented and possible causes for the issues with colour accuracy and synchronization are evaluated.

Abstract:

Since cardiovascular diseases are still a major cause of death in Germany, research in developing and improving therapies for those diseases is necessary. To achieve that, there are many possible approaches such as the usage of a simulation framework to perform cardiac simulations using the finite element method (FEM). To assist scientists and physicians in their research, the results of such simulations should be correct. This implies the necessity of a validation for the used simulation frameworks. Part of the cardiac simulation’s settings are the properties of passive material models. The parameters for those models cannot be measured directly but have to be estimated. For this work, I implemented a plugin for a cardiac simulation framework that is capable of creating slices from a whole-heart simulation as well as comparing clinical slice data. Here, human left ventricular short-axis slices, segmented from Magnetic Resonance Imaging (MRI) were used. I also developed a program for reconstructing an endocardial surface from sparse endocardial data (slice data). Subsequently, this reconstructed endocardium is also used to validate the simulation framework. The created methods are also used to estimate the passive material parameters for cardiac muscle tissue. I used synthetic slice data and clinical slice data to perform the validation of the simulation’s results. The feasibility of the parameter estimation was tested using synthetic slice data. For the validation-approach, where I compared slice data, I was able to validate the method using synthetic data. The parameter estimation using synthetic data showed that this estimation-approach is generally suitable to estimate the passive material parameters. I was also generally able to reconstruct an endocardial surface from sparse slice data. But the comparison results showed a large difference when the endocardium reconstructed from clinical data was compared to the synthetic endocardium from the simulation. This is due to disparities between the manually segmented clinical slice data and the simulation’s geometry which was based on MRI data from the same patient. Here, the clinical slice data and the simulation data from the initial state are already not matching perfectly, thus impeding the use of clinical data for the purposes of validation and parameter estimation. Using synthetic data I was able to show that my proposed methods are generally suitable for the validation of a cardiac simulation framework as well as for the passive parameter estimation. If the matching of the clinical slice data and the simulation’s geometry in the initial state can be further improved, the clinical data could also be used to estimate passive parameters (using clinical slices or an endocardium reconstructed from clinical slices). This will enable researchers to adapt existing models and methods to be more patient specific.

Abstract:

Mathematical models of the human heart contribute to understand the physiology and to support doctors in medical treatment. Hence, for applications in daily clinical routine, accurate, personalized and time efficient models are required. To make the models as simple as possible while still providing the necessary accuracy, it is important to understand the interactions of the modeling subdomains, in particular fluid dynamics, elastomechanics and electrophysiology. It is likely that data feedback loops are necessary between elastomechanics and fluid dynamics on the one side and elastomechanics and electrophysiology on the other side. Currently, connections of the subdomains are mostly neglected. In this work the fluid-structure interaction is addressed by using the pressure gradient field of an existing fluid dynamics model as an input boundary condition for an elastomechanics simulation. On the one hand, a framework extracting the pressure from the fluid dynamics is established. On the other hand, the current framework of the elastomechanics simulation is extended in order to use the pressure data. Regarding the mechano-electrical feedback, two formulations of the stretch-activated ion channel are compared with respect to the changes of calcium release and resulting action potential duration in the ventricular cell model. The results show that the deformation is influenced by the fluid dynamics feedback. Keep- ing the mean pressure in the left atrium and ventricle the same as in the initial elastome- chanics simulation, the ejection fraction remains unchanged. In the electrophysiological subdomain, the action potential duration in 0D simulations are influenced by the stretch- activated ion channels. Further, the intracellular calcium concentration is increasing.

Abstract:

Sudden, drastic changes in blood ion concentrations are common and clinically silent in patients with renal disease, from which around 42 million people in the US suffer. While otherwise healthy patients can tolerate mild hyperkalemia (i.e. high serum potassium), this condition can be life-threatening in patients suffering from cardiac disease. A non-invasive, robust, unobtrusive method for estimation of potassium and calcium concentrations in blood serum would be an important clinical advancement, providing basis for continuous monitoring an early warning for high-risk patients. During hemodialysis, patients underwent a high-resolution 12-lead electrocardiogram (ECG) recording and had 1-7 extracorporeal blood tests taken. Such data from 127 sessions of overall 42 patients was used as basis for this work. Signal processing methods were applied for signal-to-noise Ratio (SNR) improvement, followed by an averaging procedure condensing the continuous signal down to single beat ECG templates. Each template was then reduced to 14 features, whose correlations with the targets were positively assessed in previous works. Two novel features relating to the curvature of the T wave were also included. For the purpose of feature extraction a novel algorithm for finding onset and offset of the T wave was developed, which outperformed existing algorithms in a test scenario with artificially generated ECG signals. Initially, a global method for estimating potassium and calcium based on the features was developed. 112 different parametrizations of shallow neural network were tested, which all led to unsatisfactory estimation resolution.Features that exhibited the best linear dependencies with the targets out of the 14 initially chosen features were selected. Using these features and applying a 2-point calibration for each session, a simple linear model was set up, which delivered an estimation error of 0.09 mmol/l ± 0.54 mmol/l (mean ± standard deviation) for potassium and −0.01 mmol/l ± 0.12 mmol/l for calcium, after a 10-fold cross-validation. When the calibration procedure was done only with the information of the first session for each patient the estimation results were −0.2 mmol/l ± 0.68 mmol/l for potassium and −0.00 mmol/l ± 0.14 mmol/l for calcium. The methods for estimation of potassium and calcium blood ion concentration using clinical ECG data could not achieve the results of studies using simulated data. Nevertheless, the novel method presented in this work, featuring a patient-specific calibration, performs up to par with state of the art methods, exhibiting acceptable accuracy, making it clinically relevant for first-level remote monitoring of high-risk patients.

Abstract:

Als Verfahren zum Monitoring der Lungenaktivität findet die elektrische Impedanztomographie (EIT) in den vergangenenen Jahren zunehmende Anwendung. Dabei wird vergleichsweise einfach, günstig und portabel mittels Elektrodengurt um die Brust des Patienten ein Messsignal aufgenommen. Daraus kann mit Hilfe mathematischer Rekonstruktionsalgorithmen ein Bild errechnet werden, welches die Ventilationsverhältnisse in der Lunge wiedergibt. Um aus den gemessenen Daten zusätzlich das Ventilations-Perfusions-Verhältnis (VPR) anzugeben, welches als Maß für den Gasaustausch in der Lunge von hohem klinischen Interesse ist, ist zusätzlich die Auswertung der Perfusion nötig. Die Perfusion wurde dabei mittels Blutvolumenänderung, die in einer herzsynchronen Pulsatilität im EIT-Signal resultiert, bestimmt. Ziel dieser Arbeit war es, den Einfluss zu untersuchen, den verschiedene Rekonstruktionsalgorithmen und dieWahl derer Parameter - hinsichtlich dieser herzsynchronen Pulsatilität - auf das resultierende Bild haben. Dazu wurde in einem ersten Schritt ein Simulationsmodell implementiert, die Pulsatilität mittels einer kugelförmigen Impedanzänderung approximiert und anschließend eine Simulation durchgeführt. Um die erzielten Ergebnisse adäquat vergleichen zu können, wurden neben mehreren Simulationsszenarien auch die Auswertungskriterien, anhand derer die Rekonstruktionsergebnisse quantifiziert werden können, entwickelt. Dabei wurde zuerst der statische Fall eingehender betrachtet, mit dem die wichtige Wahl des korrekten Hyperparameters durchgeführt wird. Davon ausgehend wurden die Rekonstruktionseinflüsse im dynamischen Fall untersucht. Sowohl die Variation der Gefäßgröße als auch die Änderung der Leitfähigkeiten eines stark durchbluteten Gebietes, sowie die Einflüsse von anderen Gewebeschichten wurden dazu beleuchtet. Auch das Einbringen von Vorwissen, das genutzt werden kann, um das schlecht gestellte Rekonstruktionsproblem zu verbessern, wurde untersucht. In einem letzten Schritt wurden die untersuchten Rekonstruktionsalgorithmen auf real gemessene Daten angewendet, um den Einsatz in der Praxis zu testen. Zusammenfassend lässt sich festhalten, dass eine Rekonstruktion der herzsynchronen Pulsatilität aus simulierten Daten gut möglich war. Das Vorhandensein einer Impedanzänderung wurde eindeutig erkannt. Auch die Variation verschiedener Einflussfaktoren konnte quantitativ erfasst werden. Die Einflüsse verschiedener Rekonstruktionsalgorithmen waren dabei erkennbar. Abschließend lieferte die Anwendung auf gemessene Daten ein plausibles Rekonstruktionsergebnis. Die rekonstruierte Lungenlage stimmte gut mit dem MRT-Bild überein.

Abstract:

Electrical Impedance Tomography (EIT) is a radiation-free functional imaging modality used in biomedical applications, primarily, to obtain real-time images of the lungs. Since its invention, there has been an increasing scientific and clinical interest in this technology driven by the clinical need for monitoring and assessing the regional lung function at the bedside. One of the existing drawbacks of this measurement technique is, that it generates two-dimensional images from data which is influenced by three-dimensional parameters. One such parameter is pulmonary perfusion. This thesis aims to analyze the sensitivity of the measurements to a change in the conductivity due to perfusion at certain locations within the body. The aim is to estimate the contribution of varying conductivity on the readings and consequentially, generate a three-dimensional sensitivity profile which could help in better interpretation and understanding of the EIT. This profile can be used as a guideline by technicians, while developing new reconstruction algorithms, to correctly estimate the context of the measured signals. In order to achieve this, forward simulations were carried out on an FEM model of the human thorax. The model was generated by segmentation of MRI images. It was infused with spheres with higher conductivities at distinct locations in the lungs to simulate pulmonary perfusion. The results generated by the simulated measurements were calculated and compared to quantify the change in the output with respect to change in the position of the sphere relative to the belt. The main results that were sought out were: the sensitivity of the EIT measurements to perfusion and how the reconstructed images projected the effects of pulmonary perfusion in 3D, on the 2D reconstruction plane. The results obtained were in accordance with the hypotheses. The sensitivity analysis showed a varying profile which showed that the sensitivity reduced as the perfusion occurred higher up in the lung. The expected effects were also visible in the reconstructed images, which showed the position error as well as amplitude response to the perfusion changes.

Abstract:

This topic was embedded in the project of measuring blood flow intraoperativly with an optical system (surgical microscope). Measuring blood flow is important to assess the tissues perfusion. In case of hypoperfusion, the function of the tissue cannot be maintained and irreversible damages are possible. To visualize the flow dynamics in the blood, the fluores- cence dye Indocyanine Green (ICG) is injected intravenously and the dyes’ fluorescence is captured with a NIR camera. The performance of methods which assess the blood flow dynamics cannot be determined on living subjects. To facilitate the performance testing, a blood flow phantom was set up within the scope of this thesis. The main goals of this thesis were the design and the installation of the injection mimicking setup and its validation. Since no detailed flow profile of the fluids’ dynamic was available, an in-silico simulation of the flow phantom was implemented via COMSOL Multiphysics. In comparison to previous theses, the focus was on the diffusion and and convection processes of the fluorescence dye for different dimensioning of the ICG dyed bolus. The spatio-temporal morphology of the fluorescence signal depends with some constraint on the volume flow of the liquid media. For the design of the flow phantom, blood analogs were researched to obtain the flow behavior of whole blood, to support the fluorescence of ICG and to avoid impairment of the optical detection. With an ICG concentration of 0.005 g , l the optical detection of the ICG bolus was successful. The in-silico simulation of the flow dynamics and the distribution of the ICG dye showed similar results: the first but not the second boundary of the ICG bolus was detectable for a chosen dimensioning of the ICG dyed bolus. The technical model developed within the scope of this thesis shows promising results for further research. The propagation of the excitation and the fluorescence light can be calculated based on the calculated ICG distribution within the given geometry. With the optical part of the modeling, the distribution of ICG will be better understood and the development of methods which assess the blood flow will be supported.

Abstract:

Stroke is the third-most widespread cause of death in western countries. 87 % of them are ischemic strokes. The current standard therapy often involves the placement of a catheter into the affected artery. Simultaneously, hypothermia of the whole body is exercised, which slows down the metabolism of the entire human body. A new promising therapy is currently developed. It involves a cooling catheter that is placed into the common carotid artery. The catheter cools the blood flow into the affected area. However, during the therapy it is not possible to measure the cerebral temperatures without further harm for the patient. The temperatures therefore need to be simulated in a computational model, which includes most of the human arterial system. The currently used hemodynamic model included detailed arterial structures of the arms and the torso, which captured ~35 % percent of the model. However, these parts were not relevant for the temperature simulation in the cerebral region. The hypothesis was, that the arterial structures of the arms and the torso can be modeled in a more efficient way without losing cerebral influence during a stenosis. Therefore, an alternative for the modeling of these parts was desired. The chosen replacement for the arterial structures was the three element Windkessel (WK) model. The modeling of the WK was done using equation based modeling. The differential equations of the WK were implemented consistent with the current model using C code in s-function blocks of Simulink. The WK were then connected to the model at the corresponding segments. The replacements reduced the model by 304 parameters, which equals around a third of the former model. The adaption influenced the flow curves and the average flow into the replaced structures and in the cerebral region. The change in average flow rates was smaller than 5 %. After the adaption, the elements of the WK were optimized using a trust region reflective optimization, which was conducted in Matlab (lsqnonlin). The optimization improved the curve forms of the flows into the replaced arterial structure. Additionally, a sensitivity analysis of the WK parameters was executed. The results showed, that the resistors inside the Windkessel model had the highest influence. However, this influence was very small. In the end, the influence of a stenosis on the adapted model was executed and evaluated. The influence of a stenosis on the average flow into the replaced arterial structures did not change. In conclusion, the replacement had a neglectable influence on the hemodynamic model. The optimization improved the outcome of the flow curves and the sensitivity analysis showed, that the parameters do not significantly change the characteristics of the model. In the future, the outcome of the adaption can be further evaluated and the optimization could be further extended.

Abstract:

Pulmonary diseases are responsible for over 3 million deaths per year worldwide and more people are currently suffering from underdiagnosed pulmonary diseases. A proper diagnosis of pulmonary diseases requires an adequate respiratory monitoring. With the electrical impedance tomography (EIT) it is already possible to visualize the distribution of air inside the lungs and even the estimation of pulmonary perfusion by the use of the indicator dilution theory provides promising results. For perfusion estimation, a bolus of saline solution is injected to monitor the impedance changes caused by the higher conductivity of the injected indicator. However, this has only been shown for measurements in absence of respiration so far which requires a potentially strenuous ventilatory hold. This thesis deals with the detection of bolus signals during mechanical ventilation and their separation from respiration, which might further allow for a clinically more accepted method. The bolus detection was realized using a matched filter (MF) approach and evaluated on measured data with regard to sensitivity and specificity. Synthetic EIT signals were generated and used to analyze a separation based on frequency filtering. The separation quality was investigated using multiple error and similarity measures to compare original and separated signals. The automatic bolus detection algorithm provided satisfying results with a sensitivity over 96 % and specificity of 92 % for bolus injections with 5 % NaCl concentration. Further- more, the separation of those signals yielded correlation coefficients greater 0.94. However, morphological signal parameter differed between bolus signals before and after separation, which could lead to a misinterpretation of EIT indicator perfusion measurements.

Abstract:

Strokes are the third leading cause of death in western countries, of which 87 % are ischemic. A common treatment is the recanalization of the occluded vessel and therapeutic hypothermia using body surface cooling. The latter is assumed to have a neuroprotective effect, lowering the metabolic rate of the undersupplied tissue and preventing inflammations. However, cooling of the entire body leads to side-effects such as cardiac arrhythmia and organ damage, especially of the lungs. Selective cooling of the carotid blood is a new promising treatment that has the potential to give more control over the cooling, decreasing the side-effects. A temporal and spatial knowledge of the cerebral blood flow and temperature is a necessity for regulating the rate of cooling. Central to the cerebral flow distribution is an arterial circular structure, the circle of Willis. Its redundant collaterals have a compensatory effect in case of ischemia. It varies strongly across patients and underlies anatomical variations, ultimately influencing the compensatory capabilities. In this work, a preexisting hemodynamic model based on the transmission-line approach was extended by a parameterization of patient-specific factors. A simulated study on the cerebral flow was performed, aiming to analyze the parameter sensitivity and interrelations. Stenoses in the M1 segment of the middle cerebral artery (MCA) and the internal carotid artery have been inspected and their respective compensatory mechanisms and influences on said mechanisms were analyzed. In case of a stenosed internal carotid artery (ICA), the mean terminal flow of the anterior cerebral artery (ACA) was reduced from 1.39ml/s to 1.01ml/s, the MCAs from 2.56ml/s to 1.9ml/s. A stenosed M1 segment caused the MCAs flow to be reduced from 2.56ml/s to 0.7ml/s while slightly increasing flow in every other cerebral artery. Anatomical variations, with one or multiple vessels of the circle of Willis (CoW) missing could cause further decrease of terminal vessel flow: up to 14.7 % for a stenosed ICA (missing A1 segment of the other side) and 2.4% for a stenosed MCA (missing P1 segment of the same side and posterior communicating artery (PCoA) of the other side) in regard to the reference mean flow. In case of a stenosed M1 segment, the MCA showed mean flows between 0.18ml/s and 1.34ml/s, depending on the degree of collateralization. This corresponds to a maximum compensation of 53 %. For a stenosed ICA, the collaterals had less effect: depending on the collateralization, the mean flows of the ACA / MCA varied between 0.99ml/s / 1.86ml/s and 1.04ml/s / 1.98ml/s. The model can be personalized according to the specific patient and, in combination with a temperature model, it can be used to monitor the hypothermia of patients suffering from ischemic strokes. The monitoring enables regulating the temperature to appropriate setpoints.

Abstract:

Atrial fibrillation (AF) is the most common persistent cardiac arrhythmia. In order to gain a better understanding of cell behaviour, it is necessary to evaluate significant parameters. Studies have shown that local, reduced cardiac conduction velocity (CV) may be responsible for reentry tachycardia. For a local CV determination, a catheter stimulates the atrium locally from three sides and measures the response of the tissue. The available parameters are the time of stimulation and the detected extracellular potential by the catheter. The aim is to determine the anisotropic tissue properties, the direction of propagation and the CV in the area of the catheter. For this purpose, the behavior of the excitation wave was analyzed in computer models by varying the stimulation strength, the conductivity of the tissue and the fiber direction. The results have shown, that the initial stimulation and the shape of the excitation wave have an influence on the determination of CVs. These influences were quantified in order to establish an estimation method that determines the local CV, the anisotropy of the tissue and the direction of propagation. Under the most unfavorable simulated conditions, there is a deviation from the measured CVs to the real CVs of 4% in the propagation direction and of 13 % in the transverse propagation direction. The fibre direction differs by 5◦. These values are obtained without error compensation.

Abstract:

Stroke is the third most common cause of death in western society and its number is in- creasing due to demographic change. Mechanical thrombectomy is a standard therapy for ischemic stroke patients, where the blood clot is removed in a minimal-invasive surgery. In addition to thrombectomy, the induction of selective therapeutic hypothermia through direct blood cooling was suggested, as it provides neuroprotection to the brain. For realistic simulations of cerebral temperature distribution, the finely meshed brain model Colin27 was coupled to a pre-existing hemodynamic model. The cerebrum is separated from the brainstem and cerebellum by manual segmentation of the MRI data. The Brain Vasculature Database was used in combination with a modified region growing algorithm and three-dimensional probability maps to produce perfusion territories of the major cerebral arteries. The next step was to subdivide the regions of the major cerebral arteries into terminal regions to assign each terminal artery from the hemodynamic model to one brain region. Vascularization patterns from literature and region growing were used to identify those terminal regions. A coarser version of the brain was set up for temporal simulation in COMSOL and an ischemic stroke was simulated by entirely blocking the blood flow in the right MCA M1 seg- ment. The cooling effect in the perfusion territories of the major cerebral arteries, including the impact to the body core temperature were analyzed. The simulations show, that direct blood cooling in the right Common Carotid Artery decreases the spatial average temperature in the affected hemisphere by 1.7◦C after 30 minutes of cooling, while the contralateral hemisphere stays nearly unaffected. The decrease in systemic body is insignificant. The results of this thesis indicated that mild therapeutic hypothermia is achievable in the penumbra through a cooling catheter, while the effect on the rest of the body was minimized. The simulation help to better understand and predict how selective brain cooling can be applied in stroke therapy.

Abstract:

Combining polarization and spectraly resolved information at the same time gives a good insight about the sample. While spectral map, in one side, helps to quantify chromophores, the polarization information, on the other side, gives structural information of the sam- ple with good path length selectivity. Two separate instruments are being used in the biomedical research. In this work, we have analyzed each method separately to show the feasibility of combining them for building a compact instrument. We have found that, ap- plication of polarization imaging improves the results of computing the spectral map, spe- cially in analyzing chromophores beneath the skin such as: oxy and deoxy hemoglobin. This has been veri￿ed ￿rst by using dyes with known ratios, then with biological phantom and ￿nally with in vivo measurements. We have also computed the maps of melanin for dark and white skin ￿ngers and the results showed that a higher content is found in the dark one. Moreover, analysing the temporal change in absorbance of bacteria cultured in a petri dish veri￿ed to be useful for studying their growth rate whereas the polarization gating imaging helps in qualitative visualization of bacteria growth and division.

Abstract:

Modern operation microscopes deliver optical excellent pictures to the surgeon. Beam splitting allows for multiple viewers to the scene. However, the resulting image is limited in perspective. Additonally, preoperative data can only be shown as overlay on the surface or on additional external screens. As a solution a fully digital microscope is proposed. A mulicamerasetup is proposed to create a digital 3D clone of the operational situs. This allows for a multitude of view angles, and a complete integration of preoperative data with the observed scene. The data can be presented with an head-mounted display. Yet, a fully virtual approach is not suitable, interaction with tools and persons in the operation room has to be ensured. This work tackles this problem by proposing a mixed reality approach as an environment for the above data. Since see-through-HMDs, like the Hololens, don’t offer a solution for opaque pixels, a video see through approach was chosen. As an HMD the HTC Vive Pro was used, since it comes with 2 front cameras. The mixed reality was designed using the engine Unity and their plugins SteamVR, OpenVR and SRworks. The captured pictures of the frontcameras are displayed on a virtual plane. A switch to a fully virtual scene was implemented, in which 3D-models are created and shown in realtime. The mixed-reality was evaluated by a userstudy. A general suitability of the approach was shown, however it lacks on several important aspects. These include image quality, delay from captured pictures and spacial shift between real word and mixed reality. The problems are mostly due to the low quality frontcameras of the Vive Pro. The approach needs further improvement for usage in clinical application. A realtime creation and rendering of 3D-Data has been succesfully implemented, however reached framerates still have to be improved.

Abstract:

Electrical Impedance Tomography (EIT) is a radiation-free and non-invasive imaging method suitable for monitoring lung function at the bedside. The interest for EIT in the medical community is grounded in the potential ability to monitor not only pulmonary ventilation, but also pulmonary perfusion. EIT is already a well-established method for monitoring pulmonary ventilation and is currently in a research state for monitoring pulmonary perfusion. The successful simultaneous monitoring of both of these physiological processes would assist in the diagnosis of pulmonary diseases as well as aid in the optimization of mechanical ventilator settings during recruitment maneuvers. This is especially important for patients who’s lungs demonstrate an imbalance of pulmonary perfusion and ventilation, such as those diagnosed with Acute Respiratory Distress Syndrome (ARDS). A significant problem with monitoring pulmonary perfusion with EIT is the influence that the amount of electrical contrast between blood and the background tissue has on the sensitivity of EIT measurements. Therefore, the goal of this thesis was to analyse and quantify the effect that various realistic background tissues have on sensitivity, in response to a local blood volume change. To achieve this goal, EIT forward simulations were performed on three porcine FEM models representing various lung health states: healthy lungs, single-side ventilated lungs, and collapsed lungs. Additionally, each model was separately simulated using pulmonary meshes with a homogenous uniform conductivity or with a realistic heterogenous conductivity distribution. The FEM models were generated from the porcine CT data. Highly conductive spheres (110% of background tissue’s conductivity) were integrated into the pulmonary meshes of each state to simulate a local change in blood volume and therefore a local relative increase in conductivity. The results of the forward simulation were used to calculate the sensitivity to the blood volume changes for each model. Three expectations were formulated and investigated for the sensitivity results. The first stated that pulmonary regions in the proximity of the electrode belt should be more sensitive than regions close to the base or apex of the lung. The results of the simulations were in agreement with this expectation. The second expectation stated that, in comparing the realistic and homogenous pulmonary meshes within individual states, a general difference in spatial sensitivity distributions between the two types of pulmonary tissue mesh should be evident. Additionally the realistic models should be cumulatively less sensitive than the homogenous models. This expectation was also supported by the results. The final expectation was that regions of high conductivity (collapsed lung areas) should be less sensitive than regions of high conductivity. However, the results’ accordance with the third expectation varied from state to state. To model a clinical scenario, an additional comparison between varying background tissues was made to highlight the effect that background tissue might have on sensitivity, and therefore influence the reconstructed images viewed by a physician, evaluating lung function before and after recruitment maneuvers. It was found that by simply changing the background tissue from a well ventilated lung to a collapsed lung in a dorsal region at the electrode belt level, that there was a difference in sensitivity of about 0.0135 mV /S. In a different dorsal lung region at the electrode belt level, a change from collapsed pulmonary tissue to well ventilated pulmonary tissue caused a difference in sensitivity of about 0.0200 mV /S.

Abstract:

To support clinical analytics, knowledge of the heart’s wall stress is seen as an important parameter to analyse the failing heart. As this stress cannot be measured directly, it needs to be derived from other indicators. The deformation of the wall of the left ventricle, obtained from medical imaging methods, is widely used to estimate the stress. This is called the inverse problem of the heart mechanics. This thesis targeted the use of short-axis cine Magnetic Resonance Imaging (MRI) data to reconstruct the active tension in the left ventricle and to detect inhomogeneities in the myocardial wall. Based on the simulation framework CardioMechanics, it covered a sensitivity analysis of the inverse solver to enable accurate reconstruction of active tension and deformation of the left ventricle. The selected input parameters for the forward solver were considered to be ground truth. The input data used for the inverse solver were whole endocardial surface data. The results showed a good reconstruction of small infarction areas down to 1 % of the total number of elements. The system tolerated misalignment of fibre angles of up to 10◦. Statistical variability of up to 20 % of wall thickness as input data were successfully processed. A post-processing was introduced, as well as a border zone to improve active tension reconstruction in case of infarction areas. For preliminary assessment of active tension estimation based on clinical data, synthetic slice data were used to analyse the performance of the inverse solver. From the output of the forward simulation, slice data were generated. These data were remeshed to be processed by the inverse solver. Results of this analysis were not promising yet. Several possible approaches were proposed and discussed to improve the behaviour of the inverse solver. In summary, this thesis was used to explore the limits of the inverse solver showing promising results for the use of whole endocardial surface data and provided basics for the use of slice data.

Abstract:

Atrial fibrillation (AF) is the most common cardiac arrhythmia worldwide, inducing 1/3 of all arrhythmia-related hospital admissions. The disease is described via abnormal and irregular heart rhythm and goes hand in hand with a higher risk of stroke, heart failure and even death. Multiple studies found an interrelationship, between higher age and an increased risk of AF. Moreover the total number of AF-patients is rising yearly. The disease seems stubbornly to most pharmacological and surgical attempts of treatment, which is a great danger for patients diagnosed with AF. This could be due to the fact that the pathology behind the disease still remains poorly understood. Recent research attach a high value to the concept of a single reentrant wave with fibrillatory conduction, which has an organizing centre defined as phase singularity (PS). To further understand and analyse reentrant driver dynamics throughout different arrhythmic episodes post processing algorithms were implemented and tested on three different mesh-models. A simplified slab model and two anatomically correct, volumetric, tetrahedral meshes of the human atria, whereas one atrial model has two remodelled circular areas of fibrotic tissue. PSs were detected via searching for intersection of activation (defined by a set threshold) and recovery (derivative of membrane potential equals zero) of isosurfaces. In a 3D scroll wave the organizing centre of rotation represents a line of PSs defined as filament. Because of this, neighbouring PSs, which were detected, form a filament for every screenshot of the simulation. Spatio-temporal behaviour of detected filaments was considered by searching for intersection of filaments for the current time t and filaments detected in t+1. The path of each filament was tracked, leading to results which show filament dynamics. Because of many short living filaments the outcome was filtered by a minimum lifetime, leading to results which show only the important drivers for our study. To further investigate the area in which the driver seemed to be stable, an algorithm finding the point of highest probability for a PS to meander through this area, was developed. This point was defined as the area on the epicardium the PS passes by most often throughout the simulation, for one single reentrant driver. We assumed that this point of highest probability for finding a PS would lie in the centre of the rotation. Wherefore we detected the distance from the point of highest probability of finding PS to a point furthest away still belonging to the same rotation. This distance should represent an approximate radius of the rotation and include all mesh elements belonging to the rotation. The results of this thesis indicated that many short living filaments accrued because of wave collisions, visible throughout the simulation. Furthermore initiated drivers have been wiped out via the fibrotic tissue and no stable rotors were detected in this case, but the arrhythmia was still ongoing. Which means a single persistent rotor could not be the only reason for AF maintenance. All in all this study provides new methods to analyse the dynamics of reentrant drivers throughout different arrhythmic episodes. Especially the detection of the point of highest PS probability could lead to further studies to analyse the influence of fibrotic tissue applied to stable reentrant drivers in a preferably small area and the resulting influence to the arrhythmia.

Abstract:

Electrical Impedance Tomography (EIT) is already a well evaluated tool for monitoring ventilation. Previous work also showed a high potential for lung perfusion monitoring with Electrical Impedance Tomography. The simultaneous measurement of ventilation and perfusion would enable the opportunity of monitoring ventilation-perfusion ratios. This would improve the possibility of monitoring and guiding respiration of patients with lung dieseases, especially critical ill patients and potentially reducing ventilator induced lung injury. However, the lack of stability and spatial resolution of the currently used methods do not allow clinical application. Therefore the goal of this thesis is to evaluate two new methods to improve EIT perfusion measurements. First of all the voltage signals were analysed. Afterwards, a separation of the indicator signal was evaluated regarding its potential to improve the image reconstruction. The results showed a decrease of correlation between EIT and PET images when the voltages were Gamma fitted or sparsing was performed. However the analysis of the voltage measurements showed that they contain valuable information regarding the perfusion state of the lung. It is proposed that this information could be used to work on improvements of the reconstruction algorithm. In the second part of this thesis the estimation of a transfer function was evaluated, to calculate perfusion parameters. For estimating the transfer function the TSVD and Tikhonov regularization were used. Two different stable algorithms for the transfer function estimation were created. The final comparison showed a median correlation of 0,897 for the images which were reconstructed using the Tikhonov regularization. In comparison the TSVD led to a median of 0,865 and the previously used maximum-slope method led to a median correlation of 0,868. However none of the evaluated methods showed a measurable improvement of the perfusion estimation, which was the goal of this thesis. But a visual comparison of the reconstructed EIT images showed noticeable differences. This result indicated that a transfer function estmiation might have the potential to improve the spatial resolution. All investigated methods strongly depend on other necessary steps of the image reconstruction. The transfer function estimation depends on the heart region detection for estimating the Arterial Input Function (AIF). Another step, which affects the results is the comparison between the PET and EIT images. The image comparison again depends on the heart region detection, as the heart region is eliminated from the images. Therefore, it is proposed to work on a better heart region detection which might have the potential to improve the overall outcome of the image reconstruction process.

Abstract:

[Current prosthetic control schemes fall in roughly two categories: Classically programmed Direct Control, in which trigger signals switch between grip patterns, or Machine Lear- ning/AI based systems, in which Pattern Recognition is used to switch directly to the desired grip. Both systems have disadvantages, namely the slowness of Direct Control and the limited patient pool of Pattern Recognition based systems. This thesis aims to investigate weather the advantages of both systems can be merged in an Artificial Neural Network based approach to Direct Control. This system should at a minimum keep the robustness of Direct Control, speed up the trigger detection and provide patients the possibility to train the system with their own individual trigger signals. Eleven Neural Nets of three different Architectures are examined. The size of the Nets is constrained by having to run an a microprocessor. The training data is recorded like one would have to do for each patient resulting in a small dataset. Preprocessing with different methods is done to increase the available information for the Neural Networks. The effectiveness of these methods is also analysed. The Neural Networks are implemented in python using the tensorflow library. Their minimum size in the memory of a microprocessor and the required time of execution are calculated, in order to fit the constraint. This thesis shows, that it is possible to train Neural Networks successfully with a small dataset and that the preprocessing of data improves Neural Network performance. The investigated system however fails to match expectations. Constraints limit its perfor- mance and the overall complexity of the system and the training process, as well as the inconsistency of the results render this system not suitable for real world application.]

Abstract:

Determination of activation times (ATs) using noninvasive electrocardiographic imaging (ECGI) is a promising technique for future diagnosis in cardiology. However, recent studies showed artificial lines of block (ALBs) in AT maps, estimated from reconstructed source signals. Although a variety of different source models and estimation methods are used, few attempts have been made to compare these. For this reason, a systematic compari- son was performed using three different source models (surface transmembrane voltages (TMVs), extracellular potentials (EPs) on an epi-endocardial surface, and EPs on a pericar- dial surface). Four different estimation methods were compared (deflection-based temporal (DefB-T), deflection-based spatiotemporal (DefB-St), correlation-based temporal (CorrB-T), and correlation-based spatiotemporal (CorrB-St)). Four physiological cases with different pacings and 10 pathological cases with elongated scars and patches were taken into account. Monodomain simulations were performed and resulting body surface potentials (BSPs) were calculated and corrupted with an additive white Gaussian noise (AWGN) to obtain a signal to noise ratio (SNR) of 20 dB. Reconstructions were performed using second-order Tikhonov regularization with 5 different degrees of smoothing and the L-curve-method to analyze the influence of the regularization. Subsequent AT estimation showed that TMVs performed better than EPs and showed fewer ALBs. Spatiotemporal approaches showed fewer ALBs than purely temporal ones. Deflection-based (DefB) methods could depict scars, but showed many ALBs. Correlation- based (CorrB) methods performed better than DefB methods and did not show ALBs for TMVs, but overblurred scars for pathological cases. A modification of the method could be developed which resulted in the reproduction of scars without showing ALBs. In addition, it was shown that spatial oversmoothing in the reconstruction leads to character- istic ALBs and that this special kind of ALBs can be recreated by spatially smoothing true source signals. However, this does not explain all occurences of ALBs.