Abdominal lymph node region segmentation method based on ACKU-Net

Through ACKU-Net and comprehensive constraint technology, the accuracy and stability of CT image segmentation in the abdominal lymphatic region are solved, and high-precision lymphatic region segmentation is achieved, supporting more accurate clinical evaluation and decision-making.

CN118674689BActive Publication Date: 2025-03-28CHANGCHUN UNIV OF SCI & TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202410681815.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-29
Publication Date
2025-03-28
Estimated Expiration
2044-05-29

AI Technical Summary

Technical Problem

The prior art is difficult to fully automated and high-precision segmentation of CT images of abdominal lymphatic regions, resulting in a lack of accuracy, completeness and stability of segmentation results, affecting the accurate assessment and clinical decision-making of lymph node pathological status.

Method used

The abdominal lymphatic region segmentation method based on ACKU-Net is adopted to extract the morphological characteristics and spatial location of lymph tissues through CT technology, and use ACKU-Net to perform image segmentation, and area constraints, shape constraints, position constraints and connectivity constraints are constructed, and the segmentation results are optimized through the comprehensive loss function.

Benefits of technology

It improves the accuracy and stability of lymphatic region segmentation, ensures the accuracy and reliability of segmentation results, and supports better clinical diagnosis and treatment decisions.

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

Abstract

The present invention provides an abdominal lymphatic region segmentation method based on ACKU‑Net, which relates to the technical field of lymphatic region segmentation, and the specific steps include: S1. Taking medical images by CT technology, extracting the morphological characteristics of lymphatic tissue by medical images, and obtaining the spatial position of the lymphatic region, wherein the morphological characteristic parameters of the lymphatic tissue include the area of ​​the lymphatic region and the shape of the lymphatic region. The present invention can effectively distinguish the lymphatic region from the surrounding tissue, input the area constraint, shape constraint, position constraint and connectivity constraint of the lymphatic region into the segmentation loss function, construct a comprehensive loss function, segment the initial image by the data of the comprehensive loss function, binarize the constrained image, count the number of pixels marked as lymphatic tissue, further improve the accuracy and stability of the segmentation, combine the pixel ratio and the loss function value, and comprehensively evaluate the segmentation result, thereby ensuring the quality of lymphatic region segmentation.
Need to check novelty before this filing date? Find Prior Art