The software was the result of a collaboration between researchers at Peter the Great St. Petersburg Polytechnic University (SPbPU), the Institute of Experimental Medicine, and ITMO University, in Russia.
This technology may improve early diagnosis, promote preventive care, and ultimately enhance patients’ overall health.
The program was developed based on a discriminant analysis method that allows researchers to dissect a great amount of data more easily and create groups that share similar features. The technology combines clinical information, test results, and disease progression data, which may provide common patterns used to identify patients or particular features according to those specific characteristics.
This strategy allows clinicians to evaluate which treatments would be more likely to work for a patient or group of patients. Also, the software could predict if patients were likely — or not — to develop certain disorders or symptoms.
For instance, in a previous study researchers found that Parkinson’s patients who have very low blood levels of copper have increased chances of developing abnormal posture. “If a doctor knows about a potential threat in advance, he or she can start preparing for the treatment beforehand,” Marina Karpenko, PhD, associate professor at the biophysics department of SPbPU, said in a press release.
Built on the basis of artificial intelligence, the software has the capacity to constantly improve its predictive abilities, as Karpenko explained: “The program can be ‘trained’: the more information is uploaded in it, the more precise conclusions and recommendations it will provide,” she said.
Overall, the software may empower clinicians to diagnose Parkinson’s disease earlier, which will support prompt and targeted treatments to provide the best patient care and outcomes.
Researchers expect to make the software available in the near future, in a format that can be installed in any computer or smartphone.