Over the past 20 years, the prevalence of diabetes (type 1 and type 2) has been steadily rising worldwide. It is estimated that by 2050 over 1.3 billion people will be living with this chronic disease, which can lead to serious complications if left untreated.
To better diagnose it, researchers in Canada believe that doctors might want to pay greater attention to their patients' voices.
Scientists at Klick Labs have uncovered a surprising correlation between blood sugar levels and voice fundamental frequency. They found this out after conducting an experiment involving 505 participants, who were non-diabetic, pre-diabetic or had type 2 diabetics. Type 2 diabetes accounts for most cases of diabetes. It is caused by a reduced sensitivity of cells to insulin, a pancreatic hormone that regulates blood sugar levels.
For the purposes of their research, the scientists fitted the volunteers with continuous glucose monitors. They also recorded their voices several times a day for two weeks. They found that, as blood sugar levels rose, so did voice pitch. Specifically, a 1 mg/dL increase in glucose in the body resulted in a 0.02 Hz increase in vocal frequency. This change is imperceptible to the ear, but can be detected by sensitive audio analysis.
Surprisingly, this relationship was observed not only in people with diabetes, but also in those with normal blood sugar levels. The authors of the study, whose findings are published in the journal, Scientific Reports, believe that this phenomenon could be linked to the impact of glucose on the vocal folds (commonly known as the vocal cords). High blood sugar levels could potentially lead to subtle changes in vocal fold tension or hydration, resulting in increased pitch.
This discovery could revolutionize the way people with diabetes monitor their blood sugar levels. The researchers believe it would be possible to design a voice application that would enable diabetics to measure blood sugar levels using their voice – a far less invasive self-monitoring method than those currently available.
“(That) could change the game for the estimated 463 million people around the world living with Type 2 diabetes,” said lead author, Jaycee Kaufman, quoted in a news release.
This research adds to a growing body of evidence to suggest that voice can be a good indicator of an individual's state of health. Over the past decade, scientists have used artificial intelligence and machine learning systems to detect vocal biomarkers for a number of pathologies, including dementia, depression and even heart disease. – AFP Relaxnews