Nasopharyngeal carcinoma radiotherapy target region automatic segmentation method based on deep neural network

Abstract

The invention discloses a nasopharyngeal carcinoma radiotherapy target region automatic segmentation method based on a deep neural network, and belongs to the field of nasopharyngeal carcinoma radiotherapy target region automatic segmentation. The method comprises the following steps: acquiring CT image data of a patient, sketching a nasopharyngeal carcinoma radiotherapy target region and endangering organs based on the CT image data, and dividing the sketched CT image data into a training set and a verification set; constructing a deep neural network segmentation model, preprocessing trainingsamples in the training set, using the preprocessed training samples for training the deep neural network segmentation model, wherein preprocessing comprises the steps of reducing the numerical rangeof the training samples to a specified range and augmenting the training samples; and reducing the numerical range of the training samples in the verification set to a specified range, inputting thenumerical range into the trained deep neural network segmentation model, and quantitatively evaluating the recognition effect of the model. According to the invention, the model can automatically output the segmentation result of the nasopharyngeal carcinoma radiotherapy target area and organs endangering the nasopharyngeal carcinoma radiotherapy target area in a short time only by inputting the CT image data of the patient.

Description

technical field [0001] The invention relates to the field of automatic segmentation of a radiotherapy target area of ​​nasopharyngeal carcinoma, in particular to an automatic segmentation method of a radiotherapy target area of ​​nasopharyngeal carcinoma based on a deep neural network. Background technique [0002] Nasopharyngeal carcinoma is a malignant tumor formed in the nasopharynx, which is one of the most common cancers in China. According to statistics, from 2008 to 2012, the incidence rate of male nasopharyngeal cancer patients in my country was about 7.5 per 100,000 people, and the incidence rate of female patients was about 2.8 per 100,000 people. Radiation therapy, referred to as radiotherapy, uses high-energy X-rays to destroy cancer cells or slow down their growth. Radiation therapy has become one of the main treatments for nasopharyngeal carcinoma. Before patients receive radiotherapy, they need to take computed tomography (CT) or magnetic resonance images (Ma...

AI Technical Summary

Problems solved by technology

In reality, individual differences among patients are often large, and it is often difficult to define a general template to meet the needs of delineation. These factors seriously affect the accuracy of automatic delineation

Images