Medical image scanning automatic positioning method based on deep learning

Abstract

The invention provides a medical image scanning automatic positioning method based on deep learning, and the method comprises the steps: obtaining a large number of positioning sheet images, and carrying out the random segmentation of the images into a training set, a verification set, and a test set; labeling each organ needing to be labeled in each positioning sheet image; constructing a deep learning network model, taking the training set and the verification set as input of the network model for training, and obtaining training parameters; operating the network model on the test set by using the training parameters; obtaining coordinates, width and height data and category data of a left upper point of a positioning frame of each organ needing to be examined, carrying out further post-processing on the obtained positioning frame data according to the requirements of CT and PET scanning parameters to obtain final positioning frame data of a human body part needing to be examined, and completing deep learning; and scanning the patient by using the network model which finishes deep learning, and finally obtaining an automatically positioned scanning image.

Description

technical field [0001] The invention relates to an automatic positioning method for medical image scanning. Background technique [0002] Computed Tomography (CT) scanning can obtain information on continuous layers of the human body. The helical scanning speed is fast, and it is not easy to miss lesion information. It has great flexibility in image processing and can obtain image layers in any direction. It has been widely used. It is used in three-dimensional imaging of human body, angiographic imaging, cardiac imaging, cerebral perfusion imaging and other fields. [0003] CT imaging uses X-rays to scan the parts of the human body for inspection. The detector receives the X-rays after passing through the human body and converts them into visible light, which is converted into electrical signals by photoelectric converters, and then converted into digital signals by analog-to-digital converters. After the computer image reconstruction program is reconstructed into a CT ima...

AI Technical Summary

Problems solved by technology

This method is mainly aimed at delineating and classifying the target area of ​​tumor radiotherapy and organs at risk, and cannot realize the precise coordinate positioning of multiple organs and multiple parts.

Images