The voice, that unique timbre that each of us possesses, is full of information: apart from those on moods, it can also reveal precious medical data. All things considered, it is a real diagnostic tool. And I'm not talking about telemedicine or remote consultations, but about something much more "deep". The scientists of Click Labs have found that artificial intelligence can analyze vocal nuances imperceptible to the human ear to diagnose type 2 diabetes in less than a blink of an eye.
The experimental phase
The research, published in Mayo Clinic Proceedings: Digital Health (I link it to you here), involved 267 individuals, some of whom had type 2 diabetes. These participants recorded sentences on their smartphones six times a day for two weeks.
In total, More than 18.000 voice recordings were analyzed, identifying over 14 different acoustic characteristics. Vocal nuances so subtle that the human ear cannot perceive them. These hidden sounds provided the clues needed for an accurate diagnosis. How much? Currently, 89%. That's already a lot.
Jaycee Kaufman, the first author of the study, notes that the research could “transform how the medical community screens for diabetes.”
Care prospects
Imagine eliminating the need for invasive, expensive and time-consuming testing. Voice technology could completely revolutionize the way medical diagnoses are made. A very interesting perspective, which must be considered.
Naturally, further studies are needed: we still have to see, for example, how AI will behave in the presence of other medical conditions that could affect the voice.
Beyond type 2 diabetes
This research highlights the extraordinary potential of voice technology in identifying type 2 diabetes and other health conditions. The next step for Klick Labs, as mentioned, will be to replicate the study and expand voice search to look for pre-diabetes, hypertension and much more.
The news follows a recent MIT discovery of a bio-implant that can adapt more easily to the body, facilitating the delivery of drugs such as insulin.
In a world where technology and medicine are merging ever more closely, diagnosing diabetes may soon become as simple as talking on the phone.