2023-05-18 20:54:17
Scientists trained an artificial intelligence program to detect early, underlying signs of Parkinson’s disease years before the body begins to show symptoms, according to a study published in the journal Nature. American Chemical Society.
The program is called “CRANK-MS” and can be run on a laptop computer, and it detects tremors and slow movements in the body and analyzes what is happening in the neural network, which helps in predicting the disease.
The program was trained using data from blood plasma samples collected as part of a previous European-Spanish Nutrition and Cancer Study, focusing on 39 patients who developed Parkinson’s disease within 15 years of their participation in the study.
The scientists compared the patients’ data with that of 39 people who did not have the disease, enabling them to identify markers of potential importance.
The team observed that in people who had Parkinson’s disease, there was a decrease in the level of “triterpenoids” in the blood, which is an acid that regulates pressure in the body.
Thanks to the large amount of data supplied to the program, CRANK-MS was able to detect the risk of Parkinson’s disease with an accuracy of up to 96 percent.
In light of the promising results regarding Parkinson’s disease, scientists hope that this artificial intelligence program will contribute to detecting more diseases through blood samples.
Chemist William Donald, from the University of New South Wales, says the machine learning of the software has enabled it to identify the chemical markers most important in accurately predicting who will develop Parkinson’s disease in the future.
The World Health Organization defines Parkinson’s disease as a degenerative brain condition associated with motor symptoms (slow movement, tremor, stiffness and imbalance) as well as other complications, including cognitive impairment, mental health disorders, sleep disorders, pain and sensory disturbances.
Global estimates in 2019 showed that there were more than 8.5 million people suffering from Parkinson’s disease.
1684521093
#Training #detect #Parkinsons #disease
