Lung cancer TNM staging acquisition method and device, and display method

Abstract

The invention discloses a lung cancer TNM staging acquisition method. The method comprises the following steps of identifying a target nodule in a medical image; acquiring long and short paths of thetarget nodule, and generating T stages at least based on the long and short path; identifying abnormal lymph nodes in the medical image, and generating N stages by judging whether the abnormal lymph nodes have metastasis or not; acquiring whether the target nodule has distant metastasis or not, and generating M stages according to whether the target nodule has distant metastasis or not; and generating a TNM stage according to the T stage, the N stage and the M stage. According to the method, a nodule recognition model, a nodule segmentation model and a lymph node metastasis recognition model are trained, the information outputted by the models is utilized to automatically generate a T stage and an N stage, and whether distant metastasis information is generated into an M stage or not is judged in combination with doctor input or by accessing medical records and the like. According to the method, TNM staging generation efficiency is effectively improved, a doctor can conveniently and quickly know the illness state of a patient, and diagnosis efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a method and device for automatically acquiring TNM staging of lung cancer combined with image recognition and a display method for TNM staging of lung cancer. Background technique [0002] The TNM staging of intrapulmonary tumors is currently the commonly used clinical staging standard for lung cancer. The staging of lung cancer is judged comprehensively through the composition of three aspects of staging, which specifically include: used to indicate the size and location of the primary tumor, the distance from the tissue trachea and carina Far and near, whether there is invasion of pleura, chest wall, whether there is atelectasis or pneumonia, whether it invades the mediastinum or heart and great vessels, etc. T stage; used to indicate regional lymph nodes, including intrapulmonary, hilar, mediastinal and distant lymph nodes N stage for the presence or absence of metas...

AI Technical Summary

Problems solved by technology

Therefore, every time the TNM stage is determined, the doctor must personally measure the size of the nodule, record the location of the nodule, determine the relevant information, and check the TNM stage against the table. Using this method to determine the TNM stage is inefficient and error-prone

Images