Knowledge-based and data-based analysis of biomedical signals

Biomedical signal analysis is an important component in diagnosis and tracking of disease. The most common application is the use of the electrocardiogram (ECG) in the diagnosis of cardiac disorders. The ECG is a special case in that it has a predictable, repeatable pattern of known shape. Other biomedical signals, such as the electroencephalogram, present challenges due to both the lack of such a recurring pattern and the lack of specificity and physiological associations of the signal. Some basic techniques of analysis apply to all signals, which need to be supplemented to analyze signals such as the EEG. In this article, a general systems structure is presented that applies to all biomedical signals, which is then supplemented through the use of higher-order reasoning methodologies that permit both more in-depth analysis of the signal as well as the possibility for combining signal analysis results with other clinical information to provide a basis for more sophisticated decision support aids.