The use of AI for early detection of Parkinson's

Parkinson's Disease is a neurodegenerative illness that impacts an individual's movement over time, classic symptoms include tremors, rigidity, and general slowness and difficulty moving. It develops when nerve cells in a region of the brain called the substantia nigra begin to die, leading to a reduction in the chemical dopamine. Parkinson's primarily develops in older adults, typically after 60 years of age due to several age-related factors. Early detection is vital to identify so that treatment can start as soon as possible. Although there are no cures, medication can be used to help maintain functionality and quality of life for longer periods, and is most effective when started early. So how did AI help? Using AI and machine learning techniques, Matthew Shen was able to develop a model which studied the jitters and shakiness of the vocal input. Various studies have identified the advantages of voice analysis in the detection of early Parkinson's, however, the limitations of human ears mean there are often signs that are missed by physicians. Therefore, an advanced computer algorithm trained on sufficient high-quality data, as developed by Matthew Shen, could be able to detect more subtle acoustic nuances. This is exactly what he did and was able to achieve remarkable levels of accuracy using the model - approximately 90%. The speed of detection is incredibly quick and an analysis of a vocal input can be completed within a matter of seconds. His work has caused a significant advancement in our ability to detect early signs of Parkinson's whilst also drastically reducing healthcare associated costs of screening.