Researchers from the University of New South Wales Sydney and Boston University have collaboratively created an innovative AI tool that demonstrates great promise in the early detection of Parkinson’s disease. It can detect the disease even before the onset of noticeable symptoms. This groundbreaking tool utilizes neural networks to examine biomarkers present in the bodily fluids of patients.
Through this analysis, the scientists have successfully identified specific combinations of metabolites that have the potential to serve as early indicators or preventive signals for Parkinson’s disease.
Parkinson’s Disease and AI
Parkinson’s disease is a neurological disorder that results in involuntary movements, including tremors, stiffness, and challenges with balance and coordination. Symptoms typically emerge gradually and worsen as the condition advances. As the disease progresses, individuals may experience difficulties in walking and communicating.
Diagnosing Parkinson’s disease primarily relies on a comprehensive medical history, physical examination, and assessment of symptoms. Currently, there are no specific blood tests or imaging techniques available to definitively diagnose the disease. Instead, healthcare professionals rely on their expertise and clinical judgment to make an accurate diagnosis.
However, advancements in artificial intelligence (AI) hold tremendous potential in improving the detection and management of Parkinson’s disease. Researchers are exploring the use of machine learning algorithms to analyze various data sources. This also includes medical records, brain imaging, and wearable devices, to develop more accurate and objective diagnostic tools.
As part of their research, the scientists investigated blood samples obtained from healthy individuals. They did that as part of the Spanish European Prospective Investigation into Cancer and Nutrition (EPIC). The team specifically focused on a group of 39 individuals who later developed Parkinson’s disease. They got the disease even several years after the team collected the samples.
Using a machine learning program, the researchers thoroughly analyzed datasets that contained detailed information about metabolites. These are molecules that appears during the breakdown of food, drugs, or chemicals. They compared these metabolites to those of a control group comprising 39 patients who did not develop Parkinson’s. The researchers successfully identified distinct combinations of metabolites closely linked to the disease.
Diana Zhang, a researcher from UNSW, along with Associate Professor W. Alexander Donald, has introduced an advanced machine learning tool called CRANK-MS. The acronym stands for Classification and Ranking Analysis using Neural network generates Knowledge from Mass Spectrometry. It offers a significant improvement over traditional statistical methods in the analysis of metabolomics data.
What Makes CRANK-MS Different from Others?
The distinguishing feature of CRANK-MS is its ability to consider the associations and interconnectedness of metabolites. That sets it apart from conventional approaches that involve reducing the number of chemical features. In contrast to the conventional methods, Zhang and Donald opted for a different approach with CRANK-MS.
They inputted all available information into the tool without reducing the data. By employing this strategy, CRANK-MS can generate model predictions and identify key metabolites in a single step. It increases the chances of uncovering previously overlooked metabolites.
Enhancing Parkinson’s Detection Through AI Advancements
Currently, Parkinson’s disease is diagnosed based on observable physical symptoms, without the use of blood or laboratory tests for cases unrelated to genetics. However, CRANK-MS has the potential to be utilized when unusual symptoms arise, enabling the assessment of future risk for Parkinson’s disease.
It is important to note that larger validation studies are required to confirm the efficacy of CRANK-MS. Nevertheless, the initial limited study demonstrated promising outcomes, with CRANK-MS achieving an impressive accuracy rate of up to 96% in detecting Parkinson’s disease.
Additionally, the presence of polyfluorinated alkyl substances (PFAS) in those who developed the disease hints at possible exposure to industrial chemicals. These findings highlight the need for more comprehensive research to better understand the underlying mechanisms and potential environmental factors related to Parkinson’s disease.
The study’s findings present promising avenues for future research. This is particularly true for exploring the potential protective effects of triterpenoid-rich foods such as apples, olives, and tomatoes against Parkinson’s disease. This opens up opportunities to delve deeper into the role of these foods in preventing the disease and its progression. For reading other captivating tech news and valuable business tips, make sure to visit TheBusinessUp.
Interesting Related Article: “Coffee compounds may team up to fight Parkinson’s“