The researchers intend to use the sound of the voice to diagnose diseases using tools based on artificial intelligence.

The project was announced on Tuesday (13) and is funded by National Institutes of Healthwhich seeks to use the voice as a biomarker for diseases, as well as blood.

Identification of voice disorders

The study will use a human voice bank to diagnose serious diseases such as cancer and Alzheimer’s.

“The beauty of voice data is that it’s probably one of the cheapest types of data you can get from people.says Olivier Elemento, a professor at the Weill Cornell Medical Institute for Computational Biomedicine and one of the project’s principal investigators. “This is very accessible information that can be obtained from any patient..”

Research in recent years has explored the potential of the voice in diagnosing disease, but most of it has been small and isolated, says Yael Bensoussan, an otolaryngologist at USF Health and another principal investigator. There are also no large voice databases, and this is such a new area of ​​research that researchers have yet to find best practices for collecting voice data for research. “Let’s create templates for how to collect data“, she said.

The project aims to make diagnosis easier and more accessible by helping people who cannot come to see a doctor. But there are many possibilities associated with this research. This is a four-year project that could raise about $14 million in funding. He received support from the program Most2AI from the National Institutes of Health, which is engaged in projects to create ethical, accurate and accessible datasets that are used to develop AI tools.

The researchers started the project by building an app to collect voice data from participants with conditions such as vocal cord paralysis, Alzheimer’s disease, Parkinson’s disease, depression, pneumonia, and autism. They will be asked to make sounds, read sentences and complete texts through the app. At this stage of the collection, the clinician will control the action. “For example, a person with Parkinson’s disease may have a lower voice and slower speech.Bensussan says.

All collected data will be shared with other organizations and researchers. There is even interest Amazon to apply a search like this to your voice assistant Alexato help users know if they have emotional problems like depression or physical problems like a sore throat. The company is already working on patents to that effect, but the move raises several ethical and legal questions that experts are starting to think about.

However, at this early stage, the focus of research is not on creating software for home devices. The development of these tools is intended for use by physicians in offices and clinics. They are useful in resource-constrained environments where it is not possible to consult an expert. “Someone can record the patient’s voice and the app will say that there is a very high possibility of laryngeal cancer, so you should send the patient to see a specialist.Bensussan says. “AI for sorting will be very important.”

Meta, owner of Facebook, sued for collecting health data for ads
The company kept track of sensitive patient information, including information about diseases, doctors, and prescription drugs.

Via: Edge

…..