Анализ медицинских изображений с помощью нейронных сетей
Аннотация
The most frequent non-skin cancer type is breast cancer which is also named one of the deadliest diseases where early and accurate diagnosis is critical for recovery.
Nowadays, the early breast cancer’s detection can be done only by mammography Screening.
Recently, digitized tissue histopathology has become feasible to the application of computerized image analysis and machine learning techniques. In this thesis a pretrained model VGG16 with small convolutional filters has been trained using stochastic gradient descent, to classify benign and malignant breast cancer images, our model achieved 88% of sensitivity and specificity with only 12.4 % error rate. An area under the curve (AUC) value of 0.946 which is significantly good enough.
Consequently, CNNs are a promising method for this problem
Nowadays, the early breast cancer’s detection can be done only by mammography Screening.
Recently, digitized tissue histopathology has become feasible to the application of computerized image analysis and machine learning techniques. In this thesis a pretrained model VGG16 with small convolutional filters has been trained using stochastic gradient descent, to classify benign and malignant breast cancer images, our model achieved 88% of sensitivity and specificity with only 12.4 % error rate. An area under the curve (AUC) value of 0.946 which is significantly good enough.
Consequently, CNNs are a promising method for this problem