Capsule Endoscope Neural Network Lesion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The limited capacity and size of capsule endoscopes for digestive tract imaging result in high power consumption due to continuous image transmission, necessitating a reduction in power usage without compromising diagnostic accuracy.
Innovation Solution
Incorporating an artificial neural network to determine lesion areas within images generated by the capsule endoscope, which then transmits only valid images with flag bits indicating the presence of lesions, thereby reducing power consumption and storage requirements in the receiving device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous image transmission is performed, then diagnostic accuracy is maintained, but power consumption increases
Solution Approach 1:
The artificial neural network performs preliminary analysis of images locally within the capsule endoscope before transmission. By pre-determining whether lesions are present in each image frame, the system can selectively transmit only relevant images to the receiving device, thereby reducing overall power consumption while maintaining diagnostic accuracy for detected lesions.
Solution Approach 2:
The system extracts and transmits only the essential information (images with lesions) from the continuous stream of captured images. The artificial neural network filters out normal images and sends only abnormal findings to the receiving device, reducing the data transmission volume and associated power consumption while preserving diagnostic reliability.
2Loss of information
If all images are transmitted and stored, then complete diagnostic information is available, but storage requirements and processing burden increase
Solution Approach 1:
The artificial neural network extracts and transmits only images containing lesions from the complete set of captured images. This selective extraction ensures that diagnostic information related to pathological findings is preserved while significantly reducing the quantity of images requiring storage and processing at the receiving device.
Solution Approach 2:
By performing preliminary lesion detection locally before transmission, the system pre-identifies which images contain diagnostic information. This preliminary action ensures that no relevant diagnostic information is lost during the selective transmission process, while reducing the overall data volume for storage.
3Ease of operation
If the capsule endoscope size is reduced for ingestion, then ease of operation is improved, but battery capacity and processing power are limited
Solution Approach 1:
The system extracts and processes only the most critical information (lesion detection) within the constrained capsule endoscope. By using the artificial neural network to identify and transmit only relevant images, the system reduces the data transmission burden, allowing for smaller battery capacity while maintaining adequate operational duration for diagnostic purposes.
Data Source
AI summary
Provided is a capsule endoscope. The capsule endoscope includes: an imaging device configured to perform imaging on a digestive tract in vivo to generate an image; an artificial neural network configured to determine whether there is a lesion area in the image; and a transmitter configured to transmit the image based on a determination result of the artificial neural network.