Medical image-based acute gastrointestinal lesion region positioning method and system

By mapping digestive emergency image data from three-dimensional space to a one-dimensional topological sequence, and using normalized arc length coordinates and cross-validation residuals to generate responsibility scores, the problem of verifying the causal relationship between intestinal wall abnormalities and intraluminal hemorrhage in digestive emergencies was solved, and high-precision lesion localization was achieved.

CN122391177APending Publication Date: 2026-07-14THE 926TH HOSPITAL OF THE CHINESE PEOPLES LIBERATION ARMY JOINT LOGISTICS SUPPORT FORCE
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
THE 926TH HOSPITAL OF THE CHINESE PEOPLES LIBERATION ARMY JOINT LOGISTICS SUPPORT FORCE
Filing Date
2026-05-07
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Existing methods lack the technical means to verify the causal relationship between abnormal signs of the intestinal wall and the distribution of blood in the lumen when determining the source of bleeding in acute gastrointestinal events. This makes it difficult to distinguish between real bleeding foci and false signs when faced with interference from physiological congestion, inflammatory response, and imaging artifacts, thus limiting the accuracy of localization.

Method used

By acquiring multi-phase enhanced image data, extracting the intestinal centerline and calculating the normalized arc length coordinates, the image data is mapped from three-dimensional space to a one-dimensional topological sequence. The responsibility score is generated using the topological reachability constraint domain and cross-validation residuals, thereby achieving causal matching and dynamic correction between intestinal wall abnormalities and intraluminal hemorrhage.

Benefits of technology

It improves the accuracy of multi-phase information fusion, reduces the false positive rate and the missed diagnosis rate, and enables high-precision lesion localization within the critical time window of gastrointestinal emergencies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122391177A_ABST
    Figure CN122391177A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of medical image processing, in particular to a kind of based on medical image's digestive acute lesion area positioning method and system, obtain multi-phase enhancement image data;Extraction intestinal center line and calculate normalized arc length coordinate, image data is mapped as one-dimensional topological sequence relationship;Extract mixed front data and generate topological accessible constraint field, extract premonitory perfusion candidate domain in constraint field;Extract the source premonitory score and reverse support score of each candidate domain and cross check to generate residual;According to residual, extract adjacent intestinal segment auxiliary perfusion features and correct source premonitory score;Responsibility score is calculated by the logic of gate control;According to responsibility score, boundary expansion is obtained to get real responsibility lesion area;Causal cross validation can be carried out to intestinal wall abnormal sign and intraluminal hematocele distribution, effectively reduce false positive rate and missed diagnosis rate.
Need to check novelty before this filing date? Find Prior Art