???? AI identifies several subtypes of Parkinson’s: towards personalized treatments

2023-08-24 06:00:09

A team of researchers has developed an artificial intelligence capable of identifying which type of Parkinson’s disease a person suffers from. This process could eventually make it possible to envisage a personalized treatment for each patient (In the field of medicine, the term patient commonly designates a person receiving…).
Image d’illustration Pixabay

Generally affecting people over 50, Parkinson’s disease affects 25,000 new people in France each year. In all (The whole understood as the whole of what exists is often interpreted as the world or…), 8.5 million people are affected by this disease (The disease is an alteration of the functions or of the health of a living organism, animal…) neurodegenerative throughout the world (The word world can designate:). The latter affects the nerve cells of the brain (The brain is the main organ of the central nervous system of animals. The brain processes…) responsible for control (The word control can have several meanings. It can be used as a synonym for examination , de…) of our movements, which usually causes involuntary tremors, stiff muscles and slowed movements. There are different methods of treatment to date (The day or the day is the interval which separates sunrise from sunset; it is …), each presenting a very variable efficiency (In mathematics and in logic, a variable is represented by a symbol. It…) from person to person. If, in general, the processes that cause the disease are now known, doctors are not able to tell the difference according to the cases encountered. Thus, the same treatments are applied to all patients, without distinguishing any specificities. However, in a recent study whose results have just been published in the journal Nature Machine Intelligence, researchers have succeeded in developing an AI that has succeeded in distinguishing 4 subtypes of the disease, and is now able to identify which type patients are suffering from, with 95% accuracy.

To achieve this result, scientists from the Queen Square Institute of Neurology and the Francis Crick Institute generated stem cells from cells taken from patients with this disease. Subsequently, they compared these cells to a large number (The notion of number in linguistics is covered in the article “Number…) of images of stem cells, and trained the computer program (A computer program is a list of orders telling a computer what it should do. It…), via technology (The word technology has two de facto meanings:) deep learning, to distinguish certain specificities and thus identify different types of this pathology (Pathology, term coming from the ancient Greek, is literally the speech, the…).

If until now patients could lack appropriate care, the very encouraging results of this study open up the possibility of developing personalized treatments. This is a major advance in the knowledge of this disease.

1692868946
#identifies #subtypes #Parkinsons #personalized #treatments

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.