- Numerical prediction
- Numerical analysis
Disease risk prediction
Apply machine learning to large amounts of health checkup data to predict risks of diseases.
- Risks of lifestyle-related diseases within a specified time period are predicted based on individual health checkup data (e.g., height, weight, blood sugar level).
- Prediction models are trained based on health checkup data and medical prescriptions of approximately 200,000 individuals for seven years.
- Toshiba Digital Solutions Corporation has begun providing AI-based services to predict the risks of lifestyle-related diseases up to six years in the future. (in Japanese)
Benchmarks, strengths, and track record
- Achieved 96% accuracy in risk prediction for diabetes within three years.
Please include the title “Toshiba AI Technology Catalog: Disease risk prediction” or the URL in the inquiry text.
Please note that because this technology is currently the subject of R&D activities, immediate responses to inquiries may not be possible.
- Junichi Taguchi, Taihei Yamaguchi, Kenichi Doniwa, et.al.; “Development of a diabetes risk prediction algorithm using AI analysis based on health checkup data”; The 60th Annual Scientific Meeting of Japan Society of Ningen Dock, 2019.
- Joint development of AI for predicting risks of diabetes and other lifestyle-related diseases