Tumor malignancy risk stratification auxiliary diagnosis system based on artificial intelligence medical imaging

Abstract

The invention discloses an artificial intelligence medical imaging system for auxiliary diagnosis of tumor malignant risk stratification, comprising: a data acquisition module, a data preprocessing module, a model establishment module, a model verification and optimization module, a hierarchical diagnosis module and a database platform. The auxiliary diagnosis system for tumor malignant risk stratification of the present invention is based on artificial intelligence technology, which can realize successive stratification of malignant risk of tumors, simulate clinical diagnosis ideas, and detect benign lesions and malignant tumors with high precision using artificial intelligence models The ability to automatically diagnose space-occupying lesions with clear imaging features can substantially assist clinical management decision-making of space-occupying lesions, improve the existing workflow of clinical diagnosis, increase physicians' confidence in diagnosis, reduce work pressure, and reduce the risk of low malignancy The anxiety of patients with lesions has greatly improved the diagnosis rate of benign lesions and malignant tumors, and it is expected to realize the implementation of artificial intelligence clinical auxiliary diagnosis.

Description

technical field [0001] The invention relates to a system and method for auxiliary diagnosis of tumor malignancy risk stratification, in particular to a system and method for auxiliary diagnosis of tumor malignancy risk stratification of artificial intelligence medical imaging; it belongs to the technical field of artificial intelligence auxiliary diagnosis of medical imaging. Background technique [0002] Tumor refers to the new organism formed by the proliferation of local tissue cells under the action of various tumorigenic factors. Most benign lesions do not require surgical intervention without mass effect. The prognosis of early surgical intervention of malignant tumors is usually good, while the prognosis of advanced malignant tumors is generally poor whether it is surgical intervention or radiotherapy and chemotherapy. At present, medical institutions around the world are continuously investing huge sums of money in the study of the occurrence, development and treatm...

AI Technical Summary

Problems solved by technology

[0005] In general, most of the current systems use public datasets or single-center datasets for training, and the generalization and robustness of the models need to be tested; at the same time, the quality of the data is uneven, and some data even have no clear pathological diagnosis The information, data reliability and effective sample size of the data have been questioned; in addition, most of the current systems use binary classification thinking, pursuing high accuracy or high AUC (area under the ROC curve), but this classification method based on the possibility of malignancy does not Considering the actual situation of clinical application, it cannot effectively provide a reference for the clinical decision-making management of space-occupying lesions

Images