IDENTIFICATION OF ILLNESS THROUGH VOICE

Voice Diagnose Illness:

Artificial Intelligence could shortly diagnose the illness through the sound of your voice.

Voices provide a lot of information. Now, they can even help in detecting illness, researches are working on that.

The National Institute of Health is subsidizing a massive research project to collect voice data and instigate an AI that will determine the illness of people based on their speech.

Dr. Oliver Elemento is the co-principal investigator on the project.

Yeal Bensuassan, MD, is part of USF Health’s department of Otolaryngology. She is conducting an effort of gathering human voices data to be used in the identification of an illness.

Dr. Yael Bensoussan, the director of the University of South Florida's Health Voice Center said:

“When a person speaks everything from his vocal cord to breathing patterns provides latent information about health”.

Scientists could even identify depression or cancer with a human's voice. The team of researchers will start by gathering the voices of human beings with conditions in 5 different areas:

·???????Neurological disorders

·???????Voice disorders

·???????Mood disorders

·???????Respiratory disorders

·???????Pediatric disorders ( speech delay and autism)

This project was part of the program "NIH's Bridge to AI", this project was started a year ago federal government funded this project with more than $100 million. This program aimed to create a large-scale healthcare database for precision medicine.

Well, this is not the first that the voices of people have been collected by researchers and used AI to study them but on this level, this will be for the first time.

This massive project is a collaboration between USF Cornell and ten other institutions.

Bensuassan said: “We need to do something as a team and shape a network.”

As many people were seen before doing the same kind of work but all the time at a smaller level than this.

The ideal goal is to make such an app that would help transverse access to rural or underprivileged communities, by helping general physicians refer patients to specialists.

Extended, Siri or Alexa (iPhone or Android device) could be able to track down changes in the human voice, like cough, and also guide a person to seek medical treatment. It would be a worth it project once it becomes successful and also a huge revolution in the tech and AI world.

That’s a big thing so to get there, researchers must have to go ahead by amassing data, because the AI can only get as up to the mark as the database it is learning from. Researchers hope to collect more than 25, 000 voices by the end of four years with the data like clinical data and genetic information to match.

Bensuasson said:

“We want to build something scalable", she further added, just collecting data in laboratories and people coming from the educational institution to do that, then it will overthrow the purpose of the project.

There are also some barriers to this project. HIPAA it's the law that modulates medical privacy. They are not sure about researchers can carve up human voices.

Like if a person gives his voice,

Who does the voice belong to? What are researchers allowed to do with it??Can it be commercialized?

Even though health data can be separated from a person's identity and used for research, still voices are mostly detectable.

Each institution around has different rules and regulations on what can be shared or exposed which opens all kinds of ethical and legal questions that will be explored by the team of bioethicists.?