An approach for an all-embracing ECG signal analysis is introduced. Self organizing maps are applied to spatial high resolution multilead ECG (body surface ECG) recorded during the acute phase of myocardial infarction to detect characteristic changes. The algorithm deals with the measurement of similarity between different pathological signal types. In opposition to other techniques, the whole ECG-signal coded as a feature vector is used as input for the self organizing map. The results show that this approach is suitable for handling unsharp class transitions common to several problems from the medical domain.