Computer-assisted preoperative lung cancer patient N2-stage lymph node prediction system based on neural network

Abstract

The invention discloses a computer-aided system for predicting N2-stage lymph nodes of preoperative lung cancer patients based on a neural network. The prediction system comprises a first part: a data input part; the second part is a data preprocessing part; the third part is a model construction part; the fourth part is a model training part; and a fifth part: a prediction part: inputting data after preprocessing of CT features and clinical features of a to-be-predicted lung cancer patient, performing prediction by using the trained model, and outputting a prediction result: the to-be-predicted lung cancer patient is in an N2 stage or an N0 / N1 stage. The prediction system can accurately predict whether the lymph node of a T1N0M0 non-small cell lung cancer patient is in an N0 / N1 state or an N2 state, the area under curve (AUC) of a prediction result is up to 0.7847, the sensitivity is up to 89.80%, and the specificity is up to 54.50%. The prediction system plays a very important role in predicting the lifetime of a patient by a doctor, selecting an optimal treatment strategy and performing prognosis evaluation, and has a wide application prospect.

Description

technical field [0001] The invention relates to the field of preoperative N stage prediction of lung cancer patients, in particular to a computer-assisted preoperative N2 stage lymph node prediction system for lung cancer patients based on neural network. Background technique [0002] Lung cancer is a malignant tumor with the highest cancer mortality rate in the world. In a global study in 2018, it was estimated that lung cancer accounted for 11.6% of the 18.1 million new cancers, and accounted for 9.6 million deaths due to cancer. accounted for 18.4%. Among them, non-small cell lung cancer (NSCLC) accounts for about 80% of all lung cancers, about 75% of patients are found in the middle and advanced stages, and the 5-year survival rate is very low. carcinoma, adenocarcinoma, large cell carcinoma) grows and divides slowly, and spreads and metastasizes relatively late. [0003] Low-dose helical computed tomography (CT) can detect curable early-stage lung cancer, so CT has be...

AI Technical Summary

Problems solved by technology

[0007] The treatment strategies for patients with N0 / N1 and N2 mediastinal lymph nodes are different, and the prediction accuracy of the reported prediction models for patients with N2 mediastinal lymph nodes needs to be improved

Images