SMARTPHONE app technology, utilising artificial intelligence, could be used as a screening tool to aid diagnosis of Parkinson’s disease (PD) and COVID-19, according to new research lead by Dinesh Kumar, School of Engineering, Royal Melbourne Institute of Technology (RMIT) University, Australia.
Kumar and colleagues from the field of neurology and engineering developed the computerised voice assessment app using voice recordings from individuals with and without a diagnosis PD, saying ‘A’, ‘O’, and ‘M’ because these sounds lead to more accurate disease detection. These sounds were used to train the system to recognise voice differences between those with and without the disease within 10 seconds. The team also found that the technology was applicable to pulmonary diseases such as COVID-19.
Co-researcher, Quoc Cuong Ngo, also from the School of Engineering at RMIT, commented that the smartphone app the team developed is able to measure, “with great precision,” how the voice of an individual with PD or “at risk of hospitalisation from COVID-19 is different from healthy people.” Ngo further added that the technology they developed was faster and more accurate that similar artificial intelligence-based approaches.
Promising smartphone applications, such as the voice assessment app developed by Kumar and colleagues, could increase accessibility to screening worldwide and prove to be important in early detection and diagnosis of disease, resulting in earlier treatment implementation and hopefully improved outcomes. Kumar commented that implementation of this technology “could promote a community-wide screening programme, reaching people who might not otherwise seeks treatment until it’s too late.” Additionally, Kumar highlighted how this type of technology is “easy-to-use and low-cost,” can be used without the need for face-to-face contact, and enables physicians to remotely monitor the patients they serve.
Looking to the future, the research team hopes to perform a larger observational study to evaluate the technology against identifying disease progression in PD and to test the technology in assisting with diagnosis of other neurological disease.
