Ratko Magjarević

IEEE Engineering in Medicine and Biology Society

IEEE Engineering in Medicine and Biology Society (EMBS) is the world’s largest international society of Biomedical Engineers. Its 8,300 members reside in some 70 countries around the world, forming a true global connection, bringing together fascinating people, practices, information, ideas, opinion, and fellowship from one of science’s fasting growing fields. The mission of EMBS is to advance the application of engineering sciences and technology to medicine and biology, promote the profession, provide global leadership for the benefit of its members and humanity by disseminating knowledge, set standards, foster professional development, and recognize excellence.

EEG/MEG source imaging: methods, challenges, and open issues

We present the four key areas of research—preprocessing, the volume conductor, the forward problem, and the inverse problem—that affect the performance of EEG and MEG source imaging. In each key area we identify prominent approaches and methodologies that have open issues warranting further investigation within the community, challenges associated with certain techniques, and algorithms necessitating clarification of their implications. More than providing definitive answers we aim to identify important open issues in the quest of source localization.

Three-dimensional finite-element analysis of joule heating in electrochemotherapy and in vivo gene electrotransfer

Electrochemotherapy and electrogene therapy are new methods in molecular medicine based on electroporation-mediated introduction of foreign molecules (chemotherapeutic drugs, DNA) into target cells in vivo. Electrochemotherapy involves the injection of chemotherapeutic agent followed by a local delivery of a train of short high-voltage pulses to the tumor nodule (i.e. 8 square-wave pulses of 100 mus duration delivered at the repetition frequency of 1 Hz or several kHz, with a voltage-to-distance ratio of up to 1500 V/cm). For the transfer of DNA across a cell membrane a train of long low-voltage pulses (i.e. 8 rectangular pulses of 50 ms duration delivered at the repetition rate of 1 Hz, with a voltage-to-distance ratio up to 250 V/cm) is much more effective due to the electrophoretic effect on DNA molecule. In this paper we present a comprehensive analysis of tissue heating as a potential side effect of electric pulses used for electroporation-based treatments. The analysis is based on a coupled electrothermal model using 3-D finite-element approach. We studied two electrode geometries: parallel plates and a pair of needles. By setting the appropriate boundary conditions, we simulated driving of electrodes with short, high-voltage, electropermeabilizing pulses and with longer, lower voltage, electrophoretic pulses. We obtained time dependent solutions for electric field and temperature distribution by FEM solver. Based on the numerical simulations we analyzed the influence of tissue electrical conductivity and parameters of electric pulses (amplitude, duration, number of pulses, pulse repetition frequency) on the temperature distribution within the tissue and the electrodes. Results of our simulations show that at specific pulse parameters at least locally tissue heating might be significant (i.e. tissue temperatures to grow in excess of 43degC). For electrochemotherapy, this is not critical, but DNA electrotransfer may be unsuccessful due to heating-related DNA damage or denaturation.

Segmentation of hepatic vessels from MRI images for planning of electroporation-based treatments in the liver

Abstract Introduction. Electroporation-based treatments rely on increasing the permeability of the cell membrane by high voltage electric pulses delivered to tissue via electrodes. To ensure that the whole tumor is covered by the sufficiently high electric field, accurate numerical models are built based on individual patient geometry. For the purpose of reconstruction of hepatic vessels from MRI images we searched for an optimal segmentation method that would meet the following initial criteria: identify major hepatic vessels, be robust and work with minimal user input. Materials and methods. We tested the approaches based on vessel enhancement filtering, thresholding, and their combination in local thresholding. The methods were evaluated on a phantom and clinical data. Results. Results show that thresholding based on variance minimization provides less error than the one based on entropy maximization. Best results were achieved by performing local thresholding of the original de-biased image in the regions of interest which were determined through previous vessel-enhancement filtering. In evaluation on clinical cases the proposed method scored in average sensitivity of 93.68%, average symmetric surface distance of 0.89 mm and Hausdorff distance of 4.04 mm. Conclusions. The proposed method to segment hepatic vessels from MRI images based on local thresholding meets all the initial criteria set at the beginning of the study and necessary to be used in treatment planning of electroporation- based treatments: it identifies the major vessels, provides results with consistent accuracy and works completely automatically. Whether the achieved accuracy is acceptable or not for treatment planning models remains to be verified through numerical modeling of effects of the segmentation error on the distribution of the electric field.

ECG Based Prediction of Atrial Fibrillation Development Following Coronary Artery Bypass Grafting

In patients undergoing coronary artery bypass grafting (CABG) surgery, post-operative atrial fibrillation (AF) occurs with a prevalence of up to 40%. The highest incidence is seen between the second and third day after the operation. Following cardiac surgery AF may cause various complications such as hemodynamic instability, heart attack and cerebral or other thromboembolisms. AF increases morbidity, duration and expense of medical treatments. This study aims at identifying patients at high risk of post-operative AF. Early prediction of AF would provide timely prophylactic treatment and would reduce the incidence of arrhythmia. Patients at low risk of post-operative AF could be excluded on the basis of the contraindications of anti-arrhythmic drugs. The study included 50 patients in whom lead II electrocardiograms were continuously recorded for 48 h following CABG. Univariate statistical analysis was used in the search for signal features that could predict AF. The most promising ones identified were P wave duration, RR interval duration and PQ segment level. On the basis of these, a nonlinear multivariate prediction model was made by deploying a classification tree. The prediction accuracy was found to increase over time. At 48 h following CABG, the measured best smoothed sensitivity was 84.8% and the specificity 85.4%. The positive and negative predictive values were 72.7% and 92.8%, respectively, and the overall accuracy was 85.3%. With regard to the prediction accuracy, the risk assessment and prediction of post-operative AF is optimal in the period between 24 and 48 h following CABG.

A Simplified 3D Model of Whole Heart Electrical Activity and 12-Lead ECG Generation

We present a computationally efficient three-dimensional bidomain model of torso-embedded whole heart electrical activity, with spontaneous initiation of activation in the sinoatrial node, incorporating a specialized conduction system with heterogeneous action potential morphologies throughout the heart. The simplified geometry incorporates the whole heart as a volume source, with heart cavities, lungs, and torso as passive volume conductors. We placed four surface electrodes at the limbs of the torso: V R, V L, V F and V GND and six electrodes on the chest to simulate the Einthoven, Goldberger-augmented and precordial leads of a standard 12-lead system. By placing additional seven electrodes at the appropriate torso positions, we were also able to calculate the vectorcardiogram of the Frank lead system. Themodel was able to simulate realistic electrocardiogram (ECG) morphologies for the 12 standard leads, orthogonal X, Y, and Z leads, as well as the vectorcardiogram under normal and pathological heart states. Thus, simplified and easy replicable 3D cardiac bidomain model offers a compromise between computational load and model complexity and can be used as an investigative tool to adjust cell, tissue, and whole heart properties, such as setting ischemic lesions or regions of myocardial infarction, to readily investigate their effects on whole ECG morphology.

Home Care Technologies for Ambient Assisted Living

Health technology and increased medical knowledge enable accurate diagnostics and effective treatment of a large number of diseases, including those which only a decade ago were not easy to manage and cure. The interest and biomedical research in the modern society is intensively directed on disease prevention, early diagnostic and life quality improvement as well as on development of personalized healthcare especially for those chronically ill, disabled and for the aging population. The aim of the new approach in healthcare is not only to monitor and improve health of individuals, but also to increase their independence, mobility, safety and social contact through increased communication, inclusion and participation using available technologies. A large number of new medical devices for health monitoring, home care, wellness promotion, gerontotechnology, etc. have to be designed, tested and adopted to meet the special needs and demands of different population groups. These new devices for telemonitoring and telediagnostics create large amount of health related information, in most cases from sensors organized into body sensor networks. The information has to be processed, transmitted from the point of care to the healthcare system in a safe way and after managing the information in an appropriate and intelligent manner, decisions related to the persons health are to be made. This paper brings an overview of some solutions presented in literature as well as our own development of intelligent mobile monitoring devices.

Real time P-wave detector based on wavelet analysis

In ECG signals processing it is sometimes necessary to determine the timing of different ECG segments in real time, i.e. in the shortest time after they appeared in the signal. A typical example is P-wave synchronized pacing (heart stimulation synchronized with the P-wave). We used multistage methodology enabled by wavelet transform to delineate the ECG signal and develop a sensitive (98.5%) and reliable P-wave detector.

Challenges of the biomedical engineering education in Europe

Higher education in Europe has passed through a very dynamic period of changes during the last ten years. Since the signing of the Bologna Declaration in 1999 by the Ministers of Education from the EU states, European higher education system has aimed toward establishing harmonized programs enabling students and teachers to extensively exchange knowledge, ideas and skills. Education in the field of Biomedical Engineering has experienced changes also because of the research and development in the field which was more intensive than in other fields. Besides research in new power sources, it is the most intensive and productive research field. Much of the development in BME education in Europe is influenced by the European research policy expressed through the 7th Framework Programme where health is the major theme. In order to foster and support the changes in the European Higher Education Area (EHEA) according to the needs of research sector and the labor market, the Tempus scheme of projects was established. Tempus scheme aims to support the modernization of higher education and create an area of co-operation in the countries surrounding the EU. Our Tempus project, CRH-BME “Curricula Reformation and Harmonization in the field of Biomedical Engineering” aims to create guidelines for updating existing curricula in the field of BME in Europe in order to meet recent and future developments in the area, address new emerging interdisciplinary domains that appear as the result of the R&D progress and respond to the BME job market demands. In this paper, some policy and economic factors affecting BME education in Europe are discussed and the results of a BME education survey we prepared within the Tempus CHR-BME project are presented. The number of BME programmes in Europe has in the last decade significantly increased and there are more BME specializations as the result of growing complexity of the research and production in the field.

A Novel Approach to Wheeze Detection

Wheezing often accompanies pulmonary pathologies and its detection is considered of great importance for the diagnosis and management of respiratory diseases. Our aim was to develop a simple and robust algorithm for wheeze detection in respiratory sound spectra to be used for long-term monitoring and early stage assessment of asthma episode in children. The robustness of the algorithm enables wheezing detection in presence of noise and moving artifacts. Children cannot perform respiratory function tests such as peak-flow measurement and therefore we find continuous recording and processing of respiratory sounds as an alternative. The algorithm we used for wheeze detection is based on the idea of frequency domain peak detection proposed by Shabtai-Musih et al. because of its simplicity and scoring used for specifying the likelihood that the peaks in power spectra represent wheezes. In our algorithm, we have modified the way of searching peaks in the spectrogram. Before searching for peaks, wavelet denoising was used in order to remove the noise in spectrum without affecting the peaks that we were searching for. Using the scoring algorithm we were able to create a binary image of the spectrogram of the sounds - wheezes and score the length (duration) of connected components considered as wheezing. The components that did not meet length criterion were rejected and were not considered as wheezing. The algorithm was tested on respiratory sound signals from public signal databases and on our own signals recorded in a group of 26 asthmatic children. The algorithm successfully detected wheezes in all signals containing wheezing.