![Human genome view vith a focus on PARK9, one of the genes linked to Parkinson's disease.](https://www.mcknightsseniorliving.com/wp-content/uploads/sites/3/2023/05/Parkinson-Disease_G_172646513_1440X810-1024x576.jpg)
Researchers say a new machine-learning tool shows great promise in diagnosing Parkinson’s disease years before the first symptoms of the disease start appearing.
Scientists from the University of New South Wales Sydney, in collaboration with researchers from Boston University, say they have developed the new AI tool, which they say can predict the probability of Parkinson’s disease with up to 96% accuracy and up to 15 years before a clinical diagnosis, based on an analysis of chemicals in the blood. Research findings were published May 9 in the ACS Central Science journal.
Using data from a long-term health study, the researchers examined blood samples of 39 patients in whom Parkinson’s disease had developed over a period of years and compared it with a control group of 39 patients who didn’t get the disease.
UNSW researcher Diana Zhang and UNSW Associate Professor W. Alexander Donald developed a machine-learning tool called CRANK-MS, which stands for Classification and Ranking Analysis using Neural network generates Knowledge from Mass Spectrometry. Using the AI tool, the team was able to analyze metabolites in the blood that could potentially prevent or be early warning signs for Parkinson’s.
“This study is interesting at multiple levels,” Donald said in a news release. “First, the accuracy is very high for predicting Parkinson’s disease in advance of clinical diagnosis. Second, this machine learning approach enabled us to identify chemical markers that are the most important in accurately predicting who will develop Parkinson’s disease in the future.”
Donald emphasized that future studies would be needed on much larger cohorts to validate the findings, but he said the initial results were promising.
Researchers plan to make the CRANK-MS tool publicly available to other researchers or clinicians who would like to use machine learning for disease diagnosis using metabolomics data. They said the tool is easy to use and that results can be generated in less than 10 minutes on a conventional laptop.
“The application of CRANK-MS to detect Parkinson’s disease is just one example of how AI can improve the way we diagnose and monitor diseases,” Zhang said in the news release. “What’s exciting is that CRANK-MS can be readily applied to other diseases to identify new biomarkers of interest.”
Read more news from the McKnight’s Tech Daily here.