Automatic segmentation method for lung cancer radiotherapy endangered organs lung and heart

Abstract

The invention discloses an automatic segmentation method for lung cancer radiotherapy endangered organs lung and heart. The method comprises the following steps: proposing an RALP-Net segmentation algorithm, constructing a lung and heart segmentation model, segmenting a thoracic CT image of a patient by using the lung and heart segmentation model, and finally fusing and outputting segmentation results. The deep learning RALP-Net segmentation algorithm provided by the invention integrates global information and local information of the CT image by adding encoder coding and decoder decoding technologies, has the advantage of automatically and accurately segmenting the endangered organs in the lung CT image, can assist doctors in segmenting the endangered organs lung and heart in the lung cancer radiotherapy working process, and improves the efficiency of lung cancer radiotherapy. The method not only improves the segmentation accuracy, but also improves the working efficiency of doctors, and solves the problems that the current lung cancer radiotherapy endangered organ segmentation excessively depends on the subjective experience of the doctors, and the segmentation quality is uncertain due to the large workload of the doctors.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an automatic segmentation method for lungs and hearts of lung cancer radiotherapy endangered organs. Background technique [0002] Lung cancer originates from malignant tumors of the bronchial mucosa or glands of the lungs. The morbidity and mortality rates increase the fastest. It is one of the malignant tumors that threaten the health and life of the population. It has played an irreplaceable role. About 70% of lung cancer patients need radiation therapy. The fundamental purpose of radiation therapy is to use high-energy radiation to kill cancer cells in the body, and to maximize the radiation dose during the irradiation process. The target area is to kill tumor cells, and at the same time, the surrounding organs at risk are less or protected from unnecessary radiation. Since the lung area is close to important organs such as the heart, high doses of radiation may dama...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide an automatic segmentation method for lung and heart organs at risk in radiotherapy of lung cancer, which has the advantages of automatically and accurately segmenting organs at risk in CT images of the lungs, and solves the problem of relying too much on the experience of doctors in the segmentation of organs at risk in lung cancer radiotherapy. A large amount can easily lead to the problem of uncertainty in segmentation quality

Images