Benign and malignant classification method and system for digital pathological image blocks

Abstract

The invention relates to the technical field of medical images, and provides a digital pathological image block benign and malignant classification method and system, and the method comprises the steps: carrying out the preprocessing of a digital pathological section, and obtaining an image block with a corresponding label and an image block without a corresponding label; grouping the image blocks with the corresponding labels to obtain training set image blocks, verification set image blocks and test set image blocks, and performing data enhancement on the training set image blocks; constructing a deep learning model, training the deep learning model, and storing the model after training convergence to obtain a convergence model; carrying out iterative training according to the convergence model, and storing the model after training convergence; when the evaluation criteria in the verification set image blocks do not fluctuate greatly, a final model is stored; and inputting the test set image blocks into the final model, and calculating the output of the final model to obtain a benign and malignant classification result of the digital pathological image blocks. The precision and the efficiency of the classification method are greatly improved, and the effectiveness and the accuracy are ensured.

Description

technical field [0001] The invention relates to the technical field of medical images, and more particularly, to a method and system for classifying benign and malignant digital pathological image blocks. Background technique [0002] Pathological section analysis is recognized by the medical community as the gold standard for cancer diagnosis. The clinical diagnosis of many patients can only be confirmed by cytomorphological and histopathological examinations, including the region, benign and malignant nature, staging, and classification of various neoplastic diseases. Type, etc., so as to provide a basis for clinical selection of the correct treatment plan, postoperative diagnosis. At present, the diagnosis of cancer requires doctors to repeatedly check the pathological sections, which is very time-consuming and energy-consuming. Therefore, computer-aided diagnosis methods have very important clinical significance in the field of digital pathology. SUMMARY OF THE INVENT...

AI Technical Summary

Problems solved by technology

At present, the diagnosis of cancer requires doctors to repeatedly check the pathological slides, which is very time-consuming and energy-consuming

Images