Quantitative analysis result output method and apparatus of esophagus cancer lymph node metastasis

Abstract

The invention provides a quantitative analysis result output method and apparatus of esophagus cancer lymph node metastasis. The quantitative analysis result output method of esophagus cancer lymph node metastasis includes the steps: establishing a lymph node database, performing image analysis and feature recognition on the data in a training set, calculating the likelihood ratio, and estimatingthe pre-test probability; using the data after image analysis and feature recognition and the likelihood ratio to perform index screening, and using the preliminary screening data and the estimated pre-test probability to calculate the post-test probability; establishing a discrimination model according to the preliminary screening data and the post-test probability; substituting the data in the test set into the discrimination model, and if the expected goal is achieved, establishing a quantitative analysis model for esophageal cancer lymph node metastasis; if the expected goal is not achieved, correcting the discrimination model and testing again until the expected goal is achieved, and establishing a quantitative analysis model for esophageal cancer lymph node metastasis; and substituting the lymph node data of the lymph node to be analyzed into the analysis model, obtaining a quantitative analysis result of esophageal cancer lymph node metastasis of the lymph node to be analyzed through calculation and outputting the quantitative analysis result.

Description

technical field [0001] The invention relates to the field of digital medical treatment, in particular to a method and device for outputting quantitative analysis results of lymph node metastasis of esophageal cancer. Background technique [0002] Lymph node metastasis is an important factor in tumor staging, treatment selection, and prognosis. The significance and role of imaging examinations lie in the accurate discrimination and diagnosis. Currently, the most clinically researched and most widely used diagnostic criteria are usually the size of lymph nodes. It is believed that lymph nodes with larger volume (diameter) are considered to be metastatic, and lymph nodes with smaller volume (diameter) are considered non-metastatic. For lymph nodes near the critical value, it also depends on the subjective judgment of the observer. However, the diagnostic accuracy of this single index is not high, and some inflammatory enlarged lymph nodes are often misdiagnosed as metastasis, ...

AI Technical Summary

Problems solved by technology

[0003] At the same time, in addition to the medical comprehensive analysis and reasoning of images, the accuracy of traditional imaging diagnosis also depends on the doctor's knowledge structure and personal experience to varying degrees.

Moreover, some commonly used statistical methods also have some defects, such as ROC curve, etc., which can only study a single quantitative data; for descriptive data, the sensitivity and specificity of diagnosis are usually used for probability analysis one by one

Therefore, the traditional impact diagnosis results are not objective and accurate enough

Images