Abstract
In this study, high-resolution detection was aimed with an approach based on the simultaneous consideration of spatial and spectral relationships in the detection of mitotic cells in digital multispectral histopathologic images. Thanks to these studies in computer-aided diagnosis (CAD), the detection of mitotic cells which are normally difficult to examine and detect by microscopes can be performed automatically by experts. In this context, multispectral histopathologic images of mitotic and non-mitotic cells were extracted in the first step. In this case, training and test samples with the same spatial and spectral dimensions are obtained. Subsequently, classifier models were developed for different sets of training clusters using support vector machines (SVM), random forests (RF), naive Bayes (NB), and k-Nearest Neighbor classification methods. For comparison purposes, the numerical results obtained by different methods are presented in the experiments and results section.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have