Classification of breast cancer cytological specimen using convolutional neural network

13th European Workshop on Advanced Control and Diagnosis

Zejmo, Michal^1 ; Kowal, Marek^1 ; Korbicz, Józef^1 ; Monczak, Roman^2

University of Zielona Góra, Institute of Control and Computation Engineering, 65-516 Szafrana 2, Zielona Góra, Poland^1

Department of Pathomorphology, Regional Hospital in Zielona Góra, Zielona, Góra

65-046, Poland^2

关键词: Automatic classification;    Breast cancer classifications;    Classification accuracy;    Convolutional neural network;    Convolutional Neural Networks (CNN);    Gradient descent algorithms;    Microscopic image;    Neural classifiers;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/783/1/012060/pdf
DOI  :  10.1088/1742-6596/783/1/012060

来源: IOP

【 摘 要 】

The paper presents a deep learning approach for automatic classification of breast tumors based on fine needle cytology. The main aim of the system is to distinguish benign from malignant cases based on microscopic images. Experiment was carried out on cytological samples derived from 50 patients (25 benign cases + 25 malignant cases) diagnosed in Regional Hospital in Zielona Góra. To classify microscopic images, we used convolutional neural networks (CNN) of two types: GoogLeNet and AlexNet. Due to the very large size of images of cytological specimen (on average 200000 100000 pixels), they were divided into smaller patches of size 256 256 pixels. Breast cancer classification usually is based on morphometric features of nuclei. Therefore, training and validation patches were selected using Support Vector Machine (SVM) so that suitable amount of cell material was depicted. Neural classifiers were tuned using GPU accelerated implementation of gradient descent algorithm. Training error was defined as a cross-entropy classification loss. Classification accuracy was defined as the percentage ratio of successfully classified validation patches to the total number of validation patches. The best accuracy rate of 83% was obtained by GoogLeNet model. We observed that more misclassified patches belong to malignant cases.