Correct problem statements in biomedical data processing

Abstract

The correct formulation of any task appears to be not only the first step in its solution but sometimes the solution itself. Biomedical data processing is unlike other problems encountered by biologists and they often use mathematical methods without taking into consideration both the possibilities and limitations of the given methodology and the properties of the data to be processed. As a result the approach to the solution of the problem is often inadequate.Moreover, biologists encounter difficulties with the choice of a method for data conversion into the form suitable for mathematical processing. Often this leads to the incomplete utilization of the power of a given mathematical method. Sometimes only the correct data representation is sufficient to draw final conclusions. For example, data given in the form of an interval distribution histogram or mean frequency function allow the investigator to obtain information on time-dependent features of the process.Science creates and supports conditions whereby functional foundations become the field of controversy, resulting in competitive but different ways of doing the same thing. In other words the number of alternatives constantly increases due to science.In fact the problem of choice always faces the investigator not only in relation to the optimal method of data processing, but also in relation to optimally configuring electronic devices to be used in the implementation of these methods.The correctly formulated requirements for biomedical data processing could be and should be used as a basis for this choice. Thus there arises the special problem of stating correctly the processing task itself. From our point of view the correct approach to its solution must take into account all a priori information. We must systematically look at the final aims of our research (in the sense that it is a biological data processing task) and at existing methods of analysis, their scope and limitations.