Benign and malignant tumor identification method and system based on image data and deep learning

Abstract

The invention provides a medical image sequence-oriented patient-level benign and malignant tumor automatic discrimination method and system based on deep learning. The method comprises the following steps of: firstly, acquiring, sorting and marking image data of a patient, then obtaining the probability of the tumor of the patient being benign or malignant at different levels by using a tumor area detection model, a sequence classification model and an age information module, and finally, comprehensively judging whether the tumor of the patient is benign or malignant through multi-model weighted fusion. The technology is helpful for pre-judging a benign or malignant tumor based on image detection data in the early stage of a patient, so that a corresponding treatment scheme is formulated, and the prognosis effect is improved. Meanwhile, the accuracy of diagnosis of benign or malignant tumors by a doctor by using the image data in the early stage can be improved.

Description

technical field [0001] The invention relates to the technical field of target detection and recognition in pictures, and in particular to a method and system for automatically distinguishing benign and malignant tumors at a patient level based on deep learning for medical image sequences. Background technique [0002] Tumor refers to the new organism formed by the proliferation of local tissue cells under the action of various tumorigenic factors. According to the cell characteristics of new organisms and the degree of harm to the body, tumors are divided into two categories: benign tumors and malignant tumors. The vast majority of benign tumors will not become malignant, seldom recur, grow slowly, and have relatively little impact on the body, but it is not absolute. Some benign tumors still pose certain dangers to the human body, especially benign tumors that grow on vital parts of the body. will have serious consequences. Malignant tumors are more harmful to the human b...

AI Technical Summary

Problems solved by technology

In view of the low robustness of image-based tumor benign and malignant classification in the existing technology, a patient-level tumor benign and malignant discrimination method and system based on image sequences and deep learning are proposed.

Images