Computer scientists at the University of Rochester have developed an AI-powered, speech-based screening tool that can help people assess whether they are showing signs of Parkinson’s disease, the fastest growing neurological disability in the world. A study published in the journal npj Parkinson’s Disease introduces a web-based screening test that asks users to recite two pangrams—short sentences using all 26 letters of the alphabet. Within seconds, the AI analyzes the voice recordings for subtle patterns linked to Parkinson’s, with nearly 86 percent accuracy.

Study is the first to identify specific infant immune cell changes resulting from prenatal PFAS exposure, patterns associated with weaker vaccine responses, increased allergy risk, and potential autoimmunity.

A USC proof-of-concept study found that OCT imaging can measure fluid levels in the inner ear, which correlate with a patient’s degree of hearing loss. The Keck School of Medicine of USC team used the tool to scan the inner ears of 19 patients undergoing ear surgery. Six patients had normal inner ear function, four had Ménière’s disease, and nine had vestibular schwannoma (a benign tumor on a nerve that connects the inner ear to the brain). During surgery, a thick outer bone known as the mastoid was temporarily removed, allowing researchers to use OCT to collect images of the fluid compartments in the inner ear. OCT images showed that patients with Ménière’s disease or vestibular schwannoma had higher levels of a fluid called endolymph, compared to those with normal inner ear function. Increased endolymph levels were linked to greater hearing loss, indicating that measuring these fluid levels could help predict the severity of symptoms. The researchers are working to develop a smaller, more affordable version of the tool that they plan to distribute and test with surgeons and are working to adapt the technology for clinical use outside of surgery by improving the software and image-processing techniques.