IBM Announces AI Study Predicts Huntington’s Disease Progression

IBM has partnered with the CHID Foundation, an organization that aims to develop drugs that slow the progress of Huntington’s disease, and has published the results of research on artificial intelligence models that predict the onset and progression of Huntington’s disease.

This technology is an extension of the company ‘s early detection of diabetes and the use of AI to predict the occurrence of malignant breast cancer within one year with 87% accuracy . Huntington’s disease is a disease designated as an intractable disease in Japan (designated intractable disease 8). It is a progressive (degraded pathology) neurodegenerative disease in which nerve cells in the striatum in the center of the cerebrum are lost, initially using chopsticks, writing, etc. Detailed operation becomes difficult. As the condition progresses, symptoms such as movement disorders, cognitive decline, and emotional disorders (decreased ability to suppress or plan and execute emotions) appear.

IBM’s research team trains AI using brain scan images from MRI and uses signals from white matter (relatively few precedents studied in brain research) to improve cognitive and motor performance. It was measured how it changes over time.

As a result, they are “optimistic” to be able to accurately estimate functional degradation across multiple areas from a single MRI scan.

At present, there is no cure for Huntington’s disease, and it has been confirmed that the disease is likely to occur in people aged 30 to 50 years old (and can occur at all ages), but it is difficult to predict the progression of the condition. Since some symptoms can be alleviated by drugs at each stage, the prospect of “how to proceed in the future” becomes important in caring for patients.

In addition, prospects for future conditions should make it easier for clinical trial candidates to scientifically explore new drugs and treatments. There is no hope of developing a cure for the disease with AI, but it will at least help to select the right people and provide appropriate care.

Leave a Reply

Your email address will not be published. Required fields are marked *