A voice-based algorithm can predict type 2 diabetes status in USA adults: Findings from the Colive Voice study

The pressing need to reduce undiagnosed type 2 diabetes (T2D) globally calls for innovative screening approaches. This study investigates the potential of using a voice-based algorithm to predict T2D status in adults, as the first step towards developing a non-invasive and scalable screening method. We analyzed pre-specified text recordings from 607 US participants from the Colive Voice study registered on ClinicalTrials.gov (NCT04848623). Using hybrid BYOL-S/CvT embeddings, we constructed gender-specific algorithms to predict T2D status, evaluated through cross-validation based on accuracy, specificity, sensitivity, and Area Under the Curve (AUC). The best models were stratified by key factors such as age, BMI, and hypertension, and compared to the American Diabetes Association (ADA) score for T2D risk assessment using Bland-Altman analysis. The voice-based algorithms demonstrated good predictive capacity (AUC = 75% for males, 71% for females), correctly predicting 71% of male and 66% of female T2D cases. Performance improved in females aged 60 years or older (AUC = 74%) and individuals with hypertension (AUC = 75%), with an overall agreement above 93% with the ADA risk score. Our findings suggest that voice-based algorithms could serve as a more accessible, cost-effective, and noninvasive screening tool for T2D. While these results are promising, further validation is needed, particularly for early-stage T2D cases and more diverse populations.Author summary: Type 2 diabetes (T2D) is a major public health issue, affecting millions worldwide and leading to severe health complications if undiagnosed. Currently, diagnosing T2D relies on blood tests, which are invasive, costly, and challenging to implement on a large scale. This study explores a new, non-invasive approach: detecting T2D risk through voice analysis. Using data from the Colive Voice study, we developed a voice-based algorithm to predict T2D status in adults in the USA. The algorithm analyzes specific voice features and is designed to capture subtle differences in the voices of individuals with T2D compared to those without. We trained and tested the algorithm separately for men and women and observed promising results, with the algorithm showing accuracy levels comparable to traditional risk assessment tools, such as the American Diabetes Association (ADA) score. We also found that the algorithms performed better in certain subgroups, such as older women and individuals with hypertension. Our findings highlight the potential of voice analysis as an accessible and affordable screening tool for T2D, especially valuable for early detection in diverse populations and settings with limited resources. This innovative approach could transform diabetes screening by offering a practical, scalable solution for identifying those at risk.