Histopathological image classification method based on convolutional neural network

Abstract

The invention provides a histopathological image classification method based on the convolutional neural network. The method mainly comprises steps that a data set is established, a fine-tuning protocol is adopted and carried out, a pre-trained convolutional neural network is taken as a characteristic extractor and a fine-tuned CNN is taken as a classifier, for the process, firstly, images of different body parts with different texture patterns are manually selected from the scan, patch selection and accuracy calculation are carried out, secondly, a specific single-layer complete connection layer is selected for network training, thirdly, the pre-trained CNN is taken as the characteristic extractor, depth characteristics are utilized to train the linear support vector machine (SVM) for classification, and lastly, the fine-tuned CNN is taken as the classifier, and the acquired network is utilized to classify test patches. The method is advantaged in that the data set is utilized to design and train the deep network, details are richer, the data is more perfect, retrieval and classification accuracy is significantly improved, and classification of histopathological images is effectively realized.

Description

technical field [0001] The invention relates to the field of image classification, in particular to a histopathological image classification method based on a convolutional neural network. Background technique [0002] The feature analysis and classification recognition of histopathological images are the focus of histopathological image analysis research and the key to realize computer-aided disease diagnosis. The computer-aided diagnosis system combined with expert experience provides quantitative feature description and machine recognition results, which can provide quantitative and objective diagnostic basis for doctors' clinical diagnosis and review, improve the accuracy and efficiency of diagnosis, and reduce labor costs. Reduce misdiagnosis caused by differences in subjective diagnosis experience and fatigue diagnosis. The classification of histopathological images can be applied to malignant tissue and cell identification and cancer detection, to study the effects o...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem of lacking a large number of labeled histopathological scan data sets for training, the object of the present invention is to provide a histopathological image classification method based on convolutional neural network, first manually select images with different texture patterns from the scan For images of different body parts, perform patch selection and precision calculation, then select a specific single-layer fully connected layer for network training, and then use the pre-trained convolutional neural network as a feature extractor to use deep features to train for classification. A linear support vector machine, and finally a fine-tuned convolutional neural network as a classifier, using the resulting network to classify the test patches

Images