Image Generation Device for Medicine Efficacy Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating the efficacy of molecularly targeted medicines in cancer treatment do not effectively assess the therapeutic effect or the duration of effectiveness, making it difficult to determine the optimal administration period and potential need for medication switching.
Innovation Solution
An image generation device and method that utilize machine learning to generate post-administration images of lesions based on medical images, administration periods, and types of medicines, allowing for the derivation of medicine efficacy and administration stop times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If molecularly targeted medicine is continuously administered, then therapeutic efficacy is maintained, but the treatment effect weakens over time
Solution Approach 1:
The system performs preliminary action by predicting future therapeutic efficacy and administration period extension needs before the medicine actually becomes ineffective. The prediction unit forecasts when the treatment effect will weaken, allowing clinicians to plan medication switching or administration period extension in advance, preventing treatment failure rather than reacting to it
Solution Approach 2:
The system establishes feedback by continuously monitoring lesion changes through medical images and using this information to predict future treatment efficacy. The prediction results feed back into clinical decision-making, allowing dynamic adjustment of administration periods and medication switching timing based on actual treatment response rather than fixed schedules
2Measurement precision
If medical images are acquired at multiple time points to evaluate treatment effect, then accurate efficacy assessment is achieved, but examination burden and cost increase
Solution Approach 1:
The system creates virtual copies of future medical images through prediction. Instead of acquiring actual images at multiple future time points, the prediction unit generates predicted post-administration images that replicate what future images would show. This virtual copying approach maintains measurement precision while eliminating the need for repeated physical examinations
Solution Approach 2:
The system performs preliminary action by generating predicted images before actual future examinations would be needed. These predicted images provide advance information about treatment efficacy, allowing clinicians to make decisions without waiting for actual follow-up images, thereby reducing the number of required physical examinations
Data Source
AI summary
Provided are an image generation device, an image generation method, an image generation program, a learning device, a learning method, and a learning program that can check an efficacy of a medicine. A processor acquires a medical image including a lesion and information related to a type of medicine to be administered to a patient, from whom the medical image has been acquired, and an administration period of the medicine. The processor generates, from the medical image, a post-administration image indicating the lesion after the medicine is administered to the patient for the administration period.