Developed an AI tool to predict the onset of Parkinson’s disease. « Medicine in the Library

There is no blood test to identify the risk of non-genetic Parkinson’s disease, but that could change if chemists’ new machine learning toolUniversity of New South Walesin Sidney, in collaboration with Boston University will be validated.

In the research published in the journal “ACS Central Science“, the researchers described how they used neural networks to analyze biomarkers in patients’ body fluids.

The researchers examined blood samples from healthy individuals collected by the Spanish European Prospective Survey on Cancer and Nutrition (EPIC). Focusing on 39 patients who developed Parkinson’s up to 15 years later, the team ran their machine learning program on datasets containing extensive information about metabolites, the chemical compounds the body creates when it breaks down food, drugs or chemicals .

The team was able to identify unique combinations of metabolites that could prevent or potentially be early warning signs for Parkinson’s. We have entered all the information in CRANK-MS without any data reduction from the start. And from that, we can get the model prediction and identify which metabolites drive the prediction the most, all in one step.

There are no blood tests or laboratory tests to diagnose non-genetic cases. But atypical symptoms such as sleep disturbance and apathy can present in people with Parkinson’s decades before motor symptoms occur. CRANK-MS, therefore, could be used at the first sign of these atypical symptoms to rule out or rule out the risk of developing Parkinson’s in the future.

CRANK-MS is a publicly available tool for all researchers who want to use machine learning to diagnose diseases using data from metabolomics.
“Applying CRANK-MS to detect Parkinson’s disease is just one example of how AI can improve the way we diagnose and monitor disease. Interestingly, CRANK-MS can be easily applied to other diseases to identify new biomarkers of interest2.
“The tool is intuitive and on average results can be generated in less than 10 minutes on a conventional laptop.” – say the researchers.

Read the full text of the article:
Interpretable Machine Learning on Metabolomics Data Reveals Biomarkers for Parkinson’s Disease
J. Diana Zhang, Chonghua Xue, Vijaya B. Kolachalama, and William A. Donald.
ACS Cent. Sci. 2023, Publication Date:May 9, 2023

Source: University of New South Wales