Abstract:

The application of strong electrical stimuli is a common method used for terminating irregular cardiac behaviour. The study presents the influence of electrophysiological heterogeneity on the response of human hearts to electrical stimulation. The human electrophysiology was simulated using the ten Tusscher-Noble-Noble-Panfilov cell model. The anisotropic propagation of depolarisation in three-dimensional virtual myocardial preparations was calculated using bidomain equations. The research was carried out on different types of virtual cardiac wedge. The selection of the modelling parameters emphasises the influence of cellular electrophysiology on the response of the human myocardium to electrical stimulation. The simulations were initially performed on a virtual cardiac control model characterised by electrophysiological homogeneity. The second preparation incorporated the transmural electrophysiological heterogeneity characteristic of the healthy human heart. In the third model type, the normal electrophysiological heterogeneity was modified by the conditions of heart failure. The main currents responsible for repolarisation (Ito, IKs and IKI) were reduced by 25%. Successively, [Na+]i was increased by the regulation of the Na+-Ca2+ exchange function, and fibrosis was represented by decreasing electrical conductivity. Various electrical stimulation configurations were used to investigate the differences in the responses of the three different models. Monophasic and biphasic electrical stimuli were applied through rectangular paddles and needle electrodes. A whole systolic period was simulated. The distribution of the transmembrane voltage indicated that the modification of electrophysiological heterogeneity induced drastic changes during the repolarisation phase. The results illustrated that each of the heart failure conditions amplifies the modification of the response of the myocardium to electrical stimulation. Therefore a theoretical model of the failing human heart must incorporate all the characteristic features.

Abstract:

Biventricular pacing (BVP) could be improved by identifying the patient-specific optimal electrode positions. Body surface potential map (BSPM) is a non-invasive technique for obtaining the electrophysiology and pathology of a patient. The study proposes the use of BSPM as input for an automated non-invasive strategy based on a personalized computer model of the heart, to identify the patient pathology and specific optimal treatment with BVP devices. The anatomy of a patient suffering from left bundle branch block and myocardial infarction is extracted from a series of MR data sets. The clinical measurements of BSPM are used to parameterize the computer model of the heart to represent the individual pathology. Cardiac electrophysiology is simulated with ten Tusscher cell model and excitation propagation is calculated with adaptive cellular automaton, at physiological and pathological conduction levels. The optimal electrode configurations are identified by evaluating the QRS error between healthy and pathology case with/without pacing. Afterwards, the simulated ECGs for optimal pacing are compared to the post-implantation clinically measured ECGs. Both simulation and clinical optimization methods identified the right ventricular (RV) apex and the LV posterolateral regions as being the optimal electrode configuration for the patient. The QRS duration is reduced both in measured and simulated ECG after implantation with 20 and 14%, respectively. The optimized electrode positions found by simulation are comparable to the ones used in hospital. The similarity in QRS duration reduction between measured and simulated ECG signals indicates the success of the method. The computer model presented in this work is a suitable tool to investigate individual pathologies. The personalized model could assist therapy planning of BVP in patients with congestive heart failure. The proposed method could be used as prototype for further clinically oriented investigations of computerized optimization of biventricular pacing.

Abstract:

Electrode positions and timing delays influence the efficacy of biventricular pacing (BVP). Accordingly, the study focus is on BVP optimization, using a detailed three-dimensional electrophysiological model of the human heart, adapted to patient specific anatomy and pathophysiology. The research is effectuated on ten heart models with left bundle branch block and myocardial infarction derived from magnetic resonance and computer tomography data. Cardiac electrical activity is simulated with ten Tusscher cell model and adaptive cellular automaton, at physiological and pathological conduction levels. The optimization methods are based on a comparison between the electrical response of the healthy and diseased heart models, measured in terms of root mean square error (ERMS) of the excitation front and QRS duration error (EQRS). Intra- and inter-method associations of the pacing electrodes and timing delays variables were analyzed with statistical methods, i.e. t-test for dependent data, one-way ANOVA for electrode pairs and Pearson model for equivalent parameters from the two optimization methods. The results indicate that lateral left ventricle and upper or middle septal area are frequently (60% of cases) the optimal position of the left and right electrode, respectively. Statistical analysis proves that the two optimization methods are in good agreement. In conclusion, a non-invasive pre-operative BVP optimization strategy based on computer simulations can be used to identify the most beneficial patient specific electrode configuration and timing delays.

Abstract:

Fluorescence video angiography has recently been introduced in neurosurgery. Such videos are analysed by a new software to allow quantitative characterization of the blood flow during neurovascular operations (clipping an aneurysm, treatment of an angioma). For these purposes the fluorescence dye Indocyanin Green is given intravenously. After activation by a near-infrared light source the fluorescence signal is evaluated by the software. Reference measurements by using a flow phantom were performed to verify the quantitative blood flow results of the software and to validate the software algorithms. The analysis of intraoperative videos provided characteristic biological parameters allowing their implementation in the flow phantom. Under certain conditions the experiments with the help of the flow phantom showed, the results of the software parameter identification algorithmus are within the range of parameter accuracy by the reference method.