Neural CT map segmentation method based on high-precision registration

Through deep learning technology and conditional random field model, combined with multimodal image fusion and precise registration technology, the problems of low registration accuracy and lack of spatial consistency in neural CT map segmentation are solved, and high-precision and robust map segmentation effect are achieved.

CN119649040AActive Publication Date: 2025-03-18SHANGHAI YISHANG BIOTECHNOLOGY CO LTD +1
7 Cites 3 Cited by

Patent Information

Application Number
CN202510185412.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-20
Publication Date
2025-03-18
Estimated Expiration
2045-02-20

AI Technical Summary

Technical Problem

The existing neural CT map segmentation methods have problems with low registration accuracy and lack of spatial consistency, resulting in inaccurate segmentation results.

Method used

A neural CT map segmentation method based on high-precision registration is adopted to generate fusion images by acquiring neural CT images, MRI-T1 images and MRI-T2 images, and coarse registration and wavelet multi-scale fusion based on mutual information are performed. Then, deep learning technology is used to train the U-Net network for precise registration, and map segmentation is performed by combining thin plate spline deformation parameters and conditional random field model.

Benefits of technology

It improves image quality and information richness, realizes high-precision map segmentation, overcomes the limitations of traditional registration methods that are difficult to deal with complex deformation, and generates more accurate and smoother segmentation results.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of medical images, in particular to a neural CT map segmentation method based on high-precision registration. The method comprises the following steps: acquiring a nerve CT image; performing coarse registration based on mutual information according to the CT image, and performing multi-scale fusion based on wavelet to obtain a fused image; performing standard map preprocessing on a preset standard nerve CT map database according to the fused image to obtain a preprocessed standard map set; adding a multi-scale feature fusion module to a pre-established basic U-Net architecture, and embedding a space attention module to obtain a to-be-trained U-Net network; and performing deep registration network training on the to-be-trained U-Net network by using the fused image and the preprocessed standard atlas set to obtain a trained U-Net network. According to the invention, through technologies of multi-modal image fusion, deep learning registration and conditional random field optimization, a segmentation effect with higher precision and higher robustness is realized.
Need to check novelty before this filing date? Find Prior Art