Pelvic tumor CTV automatic sketching method based on deep learning

Abstract

The invention discloses a pelvic tumor CTV automatic sketching method based on deep learning, which is suitable for a pelvic lymphatic drainage area, cervical cancer CTV and rectal cancer CTV. The automatic sketching method comprises the following steps: S1, collecting CT image data and marking drainage area subareas by a clinician, and preprocessing the image data; S2, constructing a drainage area subarea deep learning segmentation model; S3, obtaining CT image data and drainage area subarea images marked by clinicians through processing in the step S1 and the step S2, inputting the data and the images into a network, training the network, and obtaining subarea contours ; S4, automatically generating a cervical cancer CTV contour through the subarea contours; and S5, automatically generating a rectal cancer CTV contour through the subarea contours. The automatic sketching method can assist doctors in sketching cervical cancer CTV and rectal cancer CTV according to patient conditions and staging conditions of patients, and an introduced dense network can effectively improve the recognition capability of pelvic lymphatic drainage areas.

Description

technical field [0001] The invention relates to the fields of medical imaging and computer technology, in particular to a method for automatically delineating pelvic lymphatic drainage areas, cervical cancer CTV and rectal cancer CTV in CT images based on a deep learning network. Background technique [0002] In the field of radiotherapy, accurate delineation of cervical cancer CTV (tumor target volume) and rectal cancer CTV has important clinical significance. During the delineation of CTV for cervical cancer and CTV for rectal cancer, clinicians must refer to the pelvic lymphatic drainage area and consider the patient's clinical stage and treatment plan for CTV delineation. In addition, when clinicians delineate the target area, they should also delineate the pelvic lymphatic drainage area. Therefore, the automatic delineation of pelvic lymphatic drainage area, cervical cancer CTV and rectal cancer CTV has very important clinical significance. [0003] At present, the me...

AI Technical Summary

Problems solved by technology

This traditional manual method has the following disadvantages: first, manual sketching consumes a lot of time, and it often takes several hours to draw a patient; second, the sketching is prone to objective errors, which are difficult to detect and easily cause medical accidents

Fourth, different doctors have different subjective understandings when considering the clinical stage and treatment plan of the same patient, resulting in inconsistencies in the sketching style

Images