CRANK-MS: A Powerful Tool to Detect Parkinson’s Disease Early

Friday - May 12 2023, 03:35 UTC - 1 year ago

In a groundbreaking new development, researchers from the University of New South Wales (UNSW) in Australia, developed a machine learning program called CRANK-MS that shows promise in detecting Parkinson's disease earlier before the onset of symptoms. This tool analyzes biomarkers in patients' bodily fluids and could be used as an early detection tool when atypical symptoms appear, either confirming or ruling out the risk of developing Parkinson's in the future.

Parkinson’s disease is a neurological disorder that affects movement. For diseases like this, early diagnosis and treatment is important.

Insert .

In a groundbreaking development, scientists from the University of New South Wales (UNSW) Sydney, in collaboration with researchers from Boston University, have unveiled a powerful new tool that shows promise in detecting Parkinson's disease before the onset of symptoms.

Published in the journal ACS Central Science, their research highlights the use of artificial intelligence (AI) and neural networks to analyze biomarkers in patients' bodily fluids.

The team examined blood samples from healthy individuals, focusing on 39 patients who later developed Parkinson's disease, the researchers utilized a machine learning program called CRANK-MS (Classification and Ranking Analysis using Neural network generates Knowledge from Mass Spectrometry) developed by UNSW researcher Diana Zhang and Associate Professor W. Alexander Donald.

Traditionally, metabolomics data is analyzed using statistical approaches. However, the team employed a different approach by considering the associations between different metabolites, leveraging the power of machine learning.

This innovative method allowed them to identify unique combinations of metabolites that could serve as potential early warning signs or even prevent the onset of Parkinson's disease.

Associate Professor Donald explains that their approach stands out from conventional methods since they used an unedited list of data. Typically, researchers reduce the number of chemical features before employing machine learning algorithms.

However, the team fed all the information into CRANK-MS without any data reduction, enabling them to obtain accurate predictions and identify the metabolites driving those predictions.

--- The significance for Parkinson’s Disease --- .

This development holds immense significance for Parkinson's disease, which is currently diagnosed by observing physical symptoms like hand tremors. There are no blood or laboratory tests available for non-genetic cases of the disease.

However, symptoms such as sleep disorders and apathy can manifest in individuals with Parkinson's long before motor symptoms appear. Therefore, CRANK-MS could be utilized as an early detection tool when these atypical symptoms emerge. It either confirms or rules out the risk of developing Parkinson's in the future.

Nevertheless, Donald emphasizes that further validation studies with larger cohorts from different regions of the world are necessary before the tool can be reliably used. In the limited cohort analyzed for this study, CRANK-MS exhibited promising results, achieving up to 96% accuracy in detecting Parkinson's disease by analyzing blood chemicals.

"This study is interesting on multiple levels," Donald said. "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 crucial in accurately predicting future Parkinson's disease. Third, some of the chemical markers driving accurate predictions have previously been implicated in cell-based assays but not in humans." .

The tool, CRANK-MS, is currently freely available and could open up the possibility to detect neurological disorders such as Parkinson's disease before the subclinical phase.