About 240 million adults worldwide live with diabetes without knowing it.
Almost 90% of diabetes cases are type 2.
The human voice can have up to 200,000 different characteristics.
One in 10 Romanians are affected by the disease of the 21st century, type 2 diabetes.
Rapid diagnosis can help keep the disease under control, and a new method could facilitate this.
A short vocal sequence could help the specialist make a diagnosis.
A voice test can recognize type 2 diabetes in 10 seconds, according to a study published in the journal Mayo Clinic Proceedings.
Voice analysis as a diagnostic tool offers many possibilities, but also major risks.
How does voice analysis work?
The human voice can have up to 200,000 different characteristics. Artificial intelligence and its algorithms can extract unique vocal patterns from the voice that match certain clinical pictures. Records of phone conversations are sufficient for analysis. Factors such as rhythm, pauses and voice pitch are examined. For example, in certain clinical pictures there are characteristic phonetic features, such as how the vowel A is pronounced for five seconds.
What are the risks of voice frequency analysis?
Proponents of voice analysis as a diagnostic tool highlight the speed and efficiency with which diseases can be identified using the voice. On the other hand, even if the tools equipped with artificial intelligence provide very accurate information, a few voice samples are not enough to actually make a well-founded diagnosis. In addition, the risk of false positives and over-diagnosis remains high. Finally, evaluation must always pass through the filter of human expertise.
2023-11-16 08:45:27
#Type #diabetes #recognized #voice #works
