Method and system for generating diaphragm lymph node segmentation model

Abstract

The invention provides a method and system for generating a longitudinal lymph node segmentation model, and relates to the technical field of medical image processing, and the method comprises the steps: obtaining lung CT images of a plurality of thoracic surgery patients, carrying out the three-dimensional reconstruction of each lung CT image, and obtaining a three-dimensional image; respectivelycarrying out three-dimensional segmentation on each three-dimensional image to obtain a three-dimensional marked image marked with a real longitudinal lymph node lesion area; grouping the three-dimensional annotation images to obtain a training set, a test set and a correction set; training the training set to obtain a diaphragm lymph node segmentation model; inputting the test set into a longitudinal lymph node segmentation model to obtain a corresponding segmentation image, and calculating the segmentation accuracy of the longitudinal lymph node segmentation model; if the segmentation accuracy is smaller than an accuracy threshold, enabling the correction set to correct the diaphragm lymph node segmentation model; and if the segmentation accuracy is not less than the accuracy threshold,storing the diaphragm lymph node segmentation model. The accuracy of longitudinal lymph node segmentation is effectively improved, manual intervention is not needed, and the practicability is high.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a method and system for generating a mediastinal lymph node segmentation model. Background technique [0002] Lung cancer has a high incidence rate, high mortality rate, and low 5-year survival rate, which is the leading cause of cancer death in the world. Lung cancer will spread and metastasize in the advanced stage, among which mediastinal lymph node metastasis is relatively common. Lung cancer has no obvious symptoms in the early stage of lymphatic metastasis, and lymph node enlargement will appear in the late stage. As the disease progresses, multiple lymph nodes swell. After lymph node metastasis of lung cancer, most of them have no good treatment, because the spread and metastasis of cancer cells is already very serious at this time, spreading all over the body through the lymphatic system, forming many new cancer lesions. Therefore, accurate segmentatio...

AI Technical Summary

Problems solved by technology

At present, the application of deep learning in clinical lung lesions is mostly limited to pulmonary nodules. However, the detection and segmentation of pulmonary mediastinal lymph nodes is of great significance to the formulation of doctors' surgical plans and lymph node dissection. A method for segmenting mediastinal lymph nodes of lung cancer on CT images of lung cancer lesions

Images