A study found that an AI model was more accurate than experienced endocrinologists in identifying adults with acromegaly based on voice recordings. The AI model correctly identified 22 out of 31 adults with acromegaly, with a higher area under the receiver operating curve compared to the endocrinologists. Voice recordings from 151 adults with acromegaly and controls were used to train the AI model, which showed a sensitivity of 71% and a specificity of 77%. The researchers believe that voice analysis has the potential to be a biomarker for various diseases, including acromegaly, but further validation studies are needed before implementation in clinical settings.
Source link
AI model using voice recognition outperforms endocrinologists for diagnosing acromegaly
