2 papers2 findings
Mild cognitive impairment
upin humans1
Voice biomarkers predict mild cognitive impairment in humans (AUC 0.963).
“The decoding analysis revealed that the combination of Task_1, 2 and 3 achieved the highest AUC performance (AUC 0.963; 100% ratio).”
Humansn = 21
AAIC 2025 abstract25 Dec 2025
Prediabetes or diabetes
upin humans1
Voice biomarkers predict prediabetes or diabetes in humans (AUC 0.85 ± 0.06, accuracy 77.9% ± 5.2%).
“In adults aged 20-30 (n = 137), a Naive Bayes model achieved the best performance, yielding 77.9% ± 5.2% accuracy, an AUC of 0.85 ± 0.06”
HumansHbA1c > 5.7% or fasting plasma glucose > 100 mg/dLn = 137
ENDO 2025 abstract22 Oct 2025
In 1,479 participants across three cohorts, a new voice taxonomy differentiated mild cognitive impairment with an effect size of d = −0.39 and tracked cognitive change.In 1,479 participants across three cohorts, a new voice taxonomy differentiated mild cognitive impairment with an effect size of d = −0.39 and tracked cognitive change.