Diagnosis From The Sound Of A Patient’s Voice

Researchers from a US start-up, Canary Speech, are working to develop technology to predict and diagnose post-traumatic stress disorder, Parkinson’s disease, concussions or even heart diseases from the sound of a patient’s voice.

There is a growing amount of evidence suggesting that many physical and mental conditions may alter the sound of a person’s voice. For example, by speaking in a more nasal tone, elongating sounds, slurring words or even sounds that are not detectable to the human ear such as jitters or a creak in the voice.

Early diagnosis can play an immense role in effectively managing conditions such as Parkinson’s and Alzheimer’s diseases, but recent studies have shown that doctors are often struggling to distinguish the symptoms. The longer the condition is undiagnosed the more symptoms become severe.

By using analytics software and big data, Canary Speech are able to identify vocal patterns that might indicate Parkinson’s or Alzheimer’s disease. This has allowed them to notice several tell-tale signs.

Another company that is developing technology to diagnose conditions with the sound of a voice is Mayo Clinic. They are teaming with Beyond Verbal, an Israeli company to test the voices of patients with coronary artery disease. There is a belief that the hardening of arteries causing chest pain may influence voice production.

In an initial study, 150 patients were asked to produce three short vocal recordings with an app developed by Beyond Verbal. After analysing the voices using the machine, 13 different vocal features associated with patients at risk of coronary artery diseases were identified.

Yuval Mor, CEO of Beyond Verbal says, “This study has been underway for two years, and it took us awhile to understand the meaning from it, but we feel like there is real substance there. Finding correlations in voice features to physical symptoms means there could be continuations in other areas”.

Mor continues to say that the research is an opportunity to develop an app that that can alert people of risks of certain conditions based on their voice.

According to Amir Lerman, a cardiologist and professor of medicine at the Mayo Clinic, this vocal test app could be a low-cost, predictive screening tool to identify patients who are at risk of heart diseases. It can also help clinicians monitor patients after cardiac surgery. For instance, changes in a patient’s voice could possibly indicate if they have stopped taking their medication.

Reference