The mitotic index (MI) is an important measure in cell proliferation studie
s. Determination of the MI is usually made by light-microscope analysis of
slide preparations. The analyst identifies and counts thousands of cells an
d reports the percentage of mitotic shapes found among the interphase nucle
i. Full automation of this process is an ambitious task, because there can
exist very few mitotic shapes among hundreds of nuclei and thousands of art
ifacts, resulting in a high probability of false positives, i.e. objects er
roneously identified as mitosis or nuclei. A semiautomated approach for MI
calculation is reported, based on the development of a neural network (NN)
far automatic identification of metaphase spreads and stimulated nuclei in
digital images of microscope preparations at 10X magnification. After segme
ntation of the objects on each image, ten different morphometrical, photome
trical and textural features are measured on each segmented object. An NN i
s used to classify the feature vectors into three classes: metaphases, nucl
ei and artifacts. The system has been able to classify correctly approximat
ely 91% of the objects in each class, in a test set of 191 mitosis, 331 nuc
lei and 387 artifacts, obtained from 30 different microscope slides, Manual
editing of false positives from the metaphase classification results allow
s the calculation of the MI with an error of 6.5%.