Bladder Endoscope AI for Hunner Lesion Position Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The lack of standardized cystoscopic criteria and insufficient knowledge among doctors leads to inaccurate diagnosis of Hunner lesions, resulting in missed diagnoses and inappropriate treatments for interstitial cystitis, particularly in regions where cystoscope observations are not routinely conducted.
Innovation Solution
A learning model is generated using endoscope image data to accurately identify Hunner lesions, utilizing deep learning techniques such as convolutional neural networks, and can be integrated into bladder endoscope systems to provide real-time position indication of Hunner lesions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If cystoscope observations are not routinely conducted in certain regions, then diagnostic criteria can be simplified and patient visits reduced, but accuracy in identifying Hunner lesions deteriorates
Solution Approach 1:
The patent replaces the mechanical/cystoscopic examination system with an AI-based image recognition system. The learned model processes endoscopic images to automatically identify Hunner lesions, eliminating the need for doctors to perform complex visual assessment and reducing reliance on routine cystoscopy while maintaining high diagnostic accuracy.
Solution Approach 2:
The patent introduces an intermediary AI model that acts as a mediator between the endoscopic image and the diagnosis. The learned model receives endoscopic images as input, processes them through neural network layers, and outputs diagnostic results, serving as an intelligent intermediary that bridges the gap between simple imaging and accurate diagnosis.
2Adaptability or versatility
If doctors lack standardized criteria and knowledge, then diagnostic flexibility is maintained, but diagnostic accuracy and consistency deteriorate
Solution Approach 1:
The patent transforms the diagnostic parameters from subjective doctor knowledge and experience to objective AI model parameters. The learned model is trained on standardized features from endoscopic images, converting qualitative diagnostic criteria into quantitative, reproducible parameters that ensure consistent and reliable diagnoses across different doctors and regions.
Solution Approach 2:
The patent creates a digital copy of diagnostic expertise through the learned model. Instead of relying on individual doctors' knowledge and experience, the model captures and replicates the diagnostic patterns of expert clinicians, making diagnostic accuracy accessible and reproducible regardless of the practitioner's level of expertise.
3Productivity
If cystoscope observations are limited to patients with bladder abnormalities, then resource utilization is optimized, but accumulation of normal bladder images is insufficient
Solution Approach 1:
The patent applies preliminary action by pre-training the learned model on a comprehensive dataset that includes both normal and abnormal bladder images before clinical use. This preliminary training phase ensures the model has sufficient exposure to normal bladder appearances, enabling it to accurately distinguish between normal and abnormal conditions without requiring extensive post-training data collection.
Data Source
AI summary
The present disclosure relates to a method for generating a learning model, a learned model, a program, and a controller of a bladder endoscope in which the program or the model is recorded. The method including: acquiring, as teaching data, endoscope image data on a Hunner lesion in a bladder; and generating the learning model by using the teaching data such that a bladder endoscope image serves as an input and the position indication of a Hunner lesion in the bladder endoscope image serves as an output. The program causing a computer to perform acquiring endoscope image data on a Hunner lesion in a bladder, inputting a target bladder endoscope image to a learning model in which a bladder endoscope image serves as an input and position-indication data on a Hunner lesion in an endoscope image serves as an output, and outputting the position indication of the Hunner lesion.