Capsule Image Review With CADe-Centered Regions of Interest
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional endoscopic and capsule endoscope systems face challenges in efficiently and accurately diagnosing gastrointestinal abnormalities due to invasive procedures, limited field of view, and high cognitive burden on diagnosticians during image review, leading to potential missed diagnoses and increased viewing time.
Innovation Solution
A method and system for processing and displaying capsule camera images using computer-assisted detection (CADe) to highlight detected abnormalities, applying lateral circle shifting to center them in the frame, and adjusting playback speed to reduce cognitive load and improve diagnostic efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If capsule endoscope systems use conventional forward-looking cameras, then the device structure is simple, but the field of view is limited and cannot see hidden areas around ridges
Solution Approach 1:
The capsule endoscope uses multiple cameras (first camera, second camera, third camera, fourth camera) positioned at different orientations to capture images from multiple directions. Each camera is segmented to cover a specific field of view, with the first and second cameras providing forward-looking views while the third and fourth cameras provide side-looking views to capture hidden areas around ridges that conventional single-camera systems cannot see.
Solution Approach 2:
The system merges images from multiple cameras with different orientations into a comprehensive panoramic view. The processing system combines the forward-looking images from the first and second cameras with the side-looking images from the third and fourth cameras to create a complete representation of the gastrointestinal tract surface, including previously invisible areas.
2Loss of time
If diagnosticians review all capsule images manually, then diagnostic accuracy is maintained, but viewing time increases and cognitive burden increases
Solution Approach 1:
The computer-assisted detection system extracts and identifies frames containing abnormalities from the complete capsule endoscope image sequence. By automatically detecting and extracting only the relevant frames with potential abnormalities, the system eliminates the need for diagnosticians to review all images manually, significantly reducing viewing time while maintaining diagnostic accuracy through automated anomaly detection.
Solution Approach 2:
The system provides feedback to the dietician by highlighting and presenting only the frames where abnormalities were detected by the computer-assisted detection algorithm. This feedback mechanism allows the dietician to focus their expertise on the most critical frames, reducing cognitive burden and viewing time while maintaining high diagnostic accuracy through the collaborative human-AI review process.
3Adaptability or versatility
If capsule cameras have sideview or panoramic view capability, then all areas of organs can be seen, but the device complexity increases
Solution Approach 1:
The capsule endoscope is designed with multi-functionality, incorporating multiple cameras that can perform both forward-looking and side-looking functions. The first and second cameras provide conventional forward viewing capability while the third and fourth cameras provide side viewing capability, allowing a single device to adapt to different viewing requirements and capture comprehensive images of the gastrointestinal tract including hidden areas.
Data Source
AI summary
The present invention provides an effective method and system for viewing a frame sequence generated from a capsule camera system. According to one method, the frame sequence comprising a picture of a GI scene covering a field of view (FoV) is received. CADe (computer-assisted detection) is applied to detect one or more target abnormalities in the frame sequence. Horizontal location of the detection area in a target CADe frame is determined. In response to the horizontal location of the detection area, circle shift laterally is applied to the target CADe frame to shift said at least one detection area to a lateral center or close to the lateral center of the target CADe frame. The CADe frames are provided.