Software for the Classification of Muscle Fibers in Histological Images.

Histochemical staining is used to characterize the types of fibers present in skeletal muscles of organisms. The identification and classification of these fibers is carried out visually by a human expert, which entails a considerable investment of resources and time. Therefore, a computational system is developed to automate the task of muscle fiber classification in a shorter time and with a high degree of efficiency. For this purpose, we use computational tools such as data mining algorithms and intelligent pattern recognition algorithms, implemented in the Java programming language, which guarantees multiplatform support. The algorithms used are K-means, Kohonen Self-Organizing Maps and an implemented expert supervised learning algorithm. The results obtained from this system can be stored in highly compatible formats, such as spreadsheets, in order to feed other systems that perform other types of analysis, for example to determine the fractal dimension of a group of fibers of the same type to distinguish whether they are randomly distributed or not. This system will be used to determine the possible patterns of distribution and organization of muscle fibers, both in normal and pathological conditions.