Body Surface Potential Mapping

What is body surface potential mapping

A Body Surface Potential Mapping (BSPM) is one of tree basic methods for investigation of heart function. Body surface potential mapping systems uses from tens to hundreds electrodes, that are placed on the thorax of a human body. First time body surface potential leads system was introduced in 1963.  Results of body surface potential measurement can be defined as four types of maps. The first type is the immediate potential map, which is created from current potential distribution on single electrodes at the given moment. The second type is the integral map defined as: This map is generated as sum in the chosen time interval. The third type is the isochronic map, in which time of detection of some characteristic points on each signal or length of some defined interval is mapped. The last type of map is differential map. It is created as subtraction of two integral maps.

Example of potential map Example of sixteen following potential maps from QRS complex interval

Using area

One of main goal is to contribute to the development of machinery that can compute the localization and geometry of a myocard infarction. This is a huge task occupying the minds of many researchers around the world. The mathematical form of this problem solution is known but a technologically feasible solution still lies in the future. Further very large and top problems is fibrillation (atrial fibrillation as well as ventricular fibrillation). A fibrillation is disorganized electrical activity of heart at that is reduced pressure of blood in body. In a case of ventricular fibrillation is big risk of sudden cardiag death.

Direct and Inverse Problems

The BSPM is surface manifestation of heart activity. The question is, how this surface manifestation computes back to heart surface. In this problematic we discriminate two cases. A situation, in which we want to find the effect of a known cause, is commonly referred to as forward (direct) problem. The situation faced everyday by physicians analyzing ECGs is the exact opposite. From an observed effect (ECG result), they have to make inferences to determine the cause (the underlying condition of the heart). In mathematical terms, this kind of problem is referred to as an inverse problem. Briefly speaking, direct problem computes the effect of a given cause (well-posed), inverse problem computes the cause of a given effect (ill-posed).

Feature extraction approach

A inverse problem solution is very difficult problematic and moreover it is deeply wedded with results of such imaging methods (Although the ECG is still the most widely-used tool for heart diagnosis, the use of other techniques that improve our ability to produce functional images of the heart, such as ultrasound and magnetic resonance imaging (MRI), is increasing rapidly). The basic research of inverse problem is done and the results of the inverse problem are useful only if to them there exists further investigation from medical imaging methods (such as MRI). The data on their own have not this advanced investigation and hence we decided to focus on BSPM feature extraction and its combination with other medical data.

Our work

Software tool

We have develop software tool for preprocessing, analysis, visualization and features extraction from ECG signal and Body Surface Potential Mapping (BSPM). We have measured data from two systems - CARDIAG 112.2 system with 80 leads and Biosemi system with 128 leads (examples of both mapping systems are shown in figures below).

Lead grid of mapping systemLead grid of mapping system

The main result of this application is integration of basic patient data (age, gender, weigh, high, systolic and diastolic pressures), information about diagnosis (anamnesis) and medication, laboratory results and measured signals. The aim of this process is obtaining of large overview about patient and to extract related information for more complex point of view. Previous description data - data originating from several different sources and usually stored in different formats - are called heterogeneous data. In medicine, the decision making process nearly always involves dealing with heterogeneous data. Despite technological advances, medical data are traditionally interpreted by physicians, based on their long training, experience and often intuition. The aim of our project is to provide novel data processing methodology in area of heart field observation and classification. Tools for long traces of synchronic heterogeneous data explanation and rapid evaluation of new data modalities will be searched by the interdisciplinary team of experts in medical engineering and medicine. In the following figures are shown examples of single application forms for enter information about diagnosis, medication and laboratory results, signal caliper and marking of characteristic points in ECG signals and the last figure is example of potential map displaying.

Form for diagnosis and medication ECG signals with caliper

ECG signals with characteristic points Surface potential map

Previous page: A-EGM Signal Processing
Next page: Cardiotocography