Algorithm for real-time binary classification of adenomas and norms images obtained by confocal microscopy
Konstantin S. Kurochka, Konstantin Panarin · 2020
Diseases of the thyroid and parathyroid gland today are a common and urgent problem. Diagnosis of these diseases is currently carried out by medical professionals by invasive intervention in order to take biopsy samples. Simplification and automation of this process by analyzing images of confocal microscopy using artificial neural networks can improve the accuracy of diagnosis and get rid of the need for invasive intervention, which makes research in this area relevant. The authors analyze various approaches and propose an optimal algorithm for solving the problem of recognition and classification of thyroid images obtained using confocal microscopy in real time.