Rare Disease Estimation Model Using Encrypted Multi-Institution Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
General doctors face challenges in making appropriate judgments on rare diseases, leading to missed opportunities for early diagnosis and treatment.
Innovation Solution
A learning device and method that utilizes secret calculation AI to construct an estimation model for rare diseases, processing patient data in an encrypted state to estimate onset probabilities, using multi-party calculation among multiple servers to learn relationships between inspection and medical care data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If patient data from multiple medical institutions is collected to improve rare disease diagnosis accuracy, then the reliability of estimation is improved, but data security and privacy protection become more difficult to ensure
Solution Approach 1:
A secure multi-party computation system acts as an intermediary between multiple medical institutions, enabling joint training of estimation models without directly sharing raw patient data. The system uses encrypted computation protocols where each institution inputs data in encrypted form, and the model training occurs in the encrypted domain, thus protecting data security while improving diagnosis accuracy through multi-institutional collaboration.
Solution Approach 2:
Instead of sharing original patient data between institutions, the system creates and shares encrypted copies or computational representations of the data. Each institution maintains its original data locally while contributing encrypted versions to the joint model training process, ensuring that no institution obtains access to other institutions' raw data while still benefiting from collective learning.
2Adaptability or versatility
If general doctors are provided with more diagnostic tools and data, then the ability to diagnose rare diseases is improved, but the complexity of the diagnostic system increases
Solution Approach 1:
The diagnostic system is segmented into two distinct components: an automated estimation model that handles complex rare disease probability calculations, and a simple user interface that presents results to doctors in an easily interpretable format. This segmentation allows the backend complexity to be hidden while providing a simple front-end experience for doctors, improving rare disease diagnostic capability without increasing perceived system complexity.
Solution Approach 2:
The estimation model automatically processes patient data and generates diagnostic probability assessments without requiring doctors to manually operate complex diagnostic tools. The system performs self-service by automatically collecting relevant data, running the estimation algorithm, and presenting results, thereby enhancing diagnostic capability while minimizing the operational burden and complexity for doctors.
Data Source
AI summary
A learning device includes a memory and processing circuitry configured to register at least inspection data and medical care data of a patient who has developed a rare disease from a plurality of medical institutions perform predetermined preprocessing on inspection data and medical care data of a patient estimate an onset probability of an estimation target patient for each of a plurality of rare diseases based on the inspection data and medical care data of the estimation target patient after the preprocessing by using an estimation model that estimates an onset probability for each of the plurality of rare diseases and use at least the inspection data and the medical care data of the patient who has developed a rare disease after the preprocessing as learning data, and cause the estimation model to learn a relationship between the inspection data and the medical care data and an onset probability.