Detection of Type 2 Diabetes from 20-second Speech Recordings: A Large-Scale Validation Study
Brann, E.; Polle, R.; Cepukaityte, G.; Georgescu, A. L.; Parsons, O.; Molimpakis, E.; Goria, S.
Show abstract
Accessible screening for type 2 diabetes (T2D) is critical, with millions of cases remaining undiagnosed globally. Here, we present the largest known real-world validation study for a speech-based T2D prediction model, trained on speech data from over 21,000 individuals, that works on features extracted from 20-second speech recordings. The model was evaluated in two stages: 1) Against self-reported diagnoses in 7,319 English-speaking participants using AUC, and 2) Against HbA1c blood tests in a subset of 801 participants drawn from the full cohort. Performance was also compared against QDiabetes and in the presence of key confounding variables. The model demonstrated clinically useful predictive capacity on self-reported data (AUC = 0.80 {+/-} 0.03), approaching QDiabetes (AUC = 0.86 {+/-} 0.03). It was robust to most demographic confounds (e.g., age and sex) and medication use, with reduced performance in the presence of comorbidities (e.g., cardiovascular disease and hypertension). At diabetes threshold of HbA1c [≥]48 mmol/mol, the model achieved an AUC of 0.75 ({+/-}0.07). This biomarker-validated speech-based tool demonstrates potential to complement existing methods through accessible, scalable screening requiring only a 20-second speech sample.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Systematic review of precision subclassification of type 2 diabetes 93%
- Estimating Heritability of Glycaemic Response to Metformin using Nationwide Electronic Health Records and Population-Sized Pedigree 91%
- Using Genomic Context Informed Genotype Data and Within-model Ancestry Adjustment to Classify Type 2 Diabetes 90%
Similar papers in this journal
- Identifying clusters of people with Multiple Long-Term Conditions using Large Language Models: a population-based study 90%
- Quantifying Device Type and Handedness Biases in a Remote Parkinson’s Disease AI-Powered Assessment 90%
- Response to Polygenic Risk: Results of the MyGeneRank Mobile Application-Based Coronary Artery Disease Study 89%
Similar papers in this journal
- Clinical interpretation of machine learning models for prediction of diabetic complications using electronic health records 95%
- Illustrating Potential Effects of Alternate Control Populations on Real-World Evidence-based Statistical Analyses 92%
- Trajectories: a framework for detecting temporal clinical event sequences from health data standardized to the OMOP Common Data Model 91%
Similar papers in this journal
- Treatment outcomes with oral anti-hyperglycaemic therapies in people with diabetes secondary to a pancreatic condition (type 3c diabetes): A population-based cohort study 92%
- Covid-19 fatality prediction in people with diabetes and prediabetes using a simple score at hospital admission 91%
- Waist circumference thresholds predicting incident dysglycemia and type 2 diabetes in Black African men and women 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.