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

被引:0
|
作者
Elbeji, Abir [1 ]
Pizzimenti, Megane [1 ]
Aguayo, Gloria [1 ]
Fischer, Aurelie [1 ]
Ayadi, Hanin [1 ]
Mauvais-Jarvis, Franck [2 ,3 ]
Riveline, Jean-Pierre [4 ,5 ,6 ]
Despotovic, Vladimir [7 ]
Fagherazzi, Guy [1 ]
机构
[1] Luxembourg Inst Hlth, Dept Precis Hlth, Deep Digital Phenotyping Res Unit, 1 A B Rue Thomas Edison, L-1445 Strassen, Luxembourg
[2] Tulane Univ, Sch Med, Deming Dept Med, Sect Endocrinol & Metab, New Orleans, LA USA
[3] VA Med Ctr, Southeast Louisiana, New Orleans, LA USA
[4] Inst Necker Enfants Malad, Inserm, Immediab Lab, CNRS,UMR 8253,U1151, Paris, France
[5] Lariboisiere Univ Hosp, Assistance Publ Hop Paris, Dept Diabetol Endocrinol & Nutr, Paris, France
[6] Univ Paris Cite, Inst Necker Enfants Malad, Immediab Lab, INSERM,UMR S1151,CNRS,UMR S8253, Paris, France
[7] Luxembourg Inst Hlth, Bioinformat Platform, 1 A B rue Thomas Edison, L-1445 Strassen, Luxembourg
来源
PLOS DIGITAL HEALTH | 2024年 / 3卷 / 12期
关键词
RISK-FACTORS; ASSOCIATION; DEPRESSION; MELLITUS; SCORE;
D O I
10.1371/journal.pdig.0000679
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
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.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] A Voice-Based AI Algorithm Can Predict Type 2 Diabetes Status-Findings from the Colive Voice Study on US Adult Participants
    Elbeji, Abir
    Pizzimenti, Megane
    Aguayo, Gloria A.
    Fischer, Aurelie
    Ayadi, Hanin
    Mauvais-Jarvis, Franck
    Riveline, Jean-Pierre
    Despotovic, Vladimir
    Fagherazzi, Guy
    DIABETES, 2024, 73
  • [2] Can we screen for type 2 diabetes using voice? Findings from the Colive Voice study
    Fagherazzi, G.
    Despotovic, V.
    Pizzimenti, M.
    Aguayo, G.
    Fischer, A.
    Ayadi, H.
    Mauvais-Jarvis, F.
    Riveline, J. -P.
    Elbeji, A.
    DIABETOLOGIA, 2024, 67 : S185 - S186
  • [3] Digital voice-based biomarker for monitoring respiratory quality of life: findings from the colive voice study
    Despotovic, Vladimir
    Elbeji, Abir
    Funfgeld, Kevser
    Pizzimenti, Megane
    Ayadi, Hanin
    Nazarov, Petr V.
    Fagherazzi, Guy
    BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2024, 96
  • [4] Readiness of adults with type 1 diabetes and diabetes caregivers for diabetes distress monitoring using a voice-based digital health solution: insights from the PsyVoice mixed methods study
    Vergara, Franchesca Aguirre
    Pinker, India
    Fischer, Aurelie
    Seuring, Till
    Tichomirowa, Maria A.
    de Beaufort, Carine
    Kamp, Siri-Maria
    Fagherazzi, Guy
    Aguayo, Gloria A.
    BMJ OPEN, 2025, 15 (01):
  • [5] IDENTIFICATION OF VOCAL BIOMARKERS FOR SCREENING DIABETES AND MONITORING HEALTH OF PEOPLE WITH DIABETES: PRELIMINARY RESULTS FROM THE COLIVE VOICE STUDY
    Elbeji, A.
    Aguayo, G.
    Fischer, A.
    Fagherazzi, G.
    DIABETES TECHNOLOGY & THERAPEUTICS, 2022, 24 : A224 - A224
  • [6] Instrumental Acoustic Voice Characteristics in Adults with Type 2 Diabetes
    Pinyopodjanard, Sittichai
    Suppakitjanusant, Pichatorn
    Lomprew, Prangorn
    Kasemkosin, Nittaya
    Chailurkit, Laor
    Ongphiphadhanakul, Boonsong
    JOURNAL OF VOICE, 2021, 35 (01) : 116 - 121
  • [7] Digital Vocal Biomarker of Smoking Status Using Ecological Audio Recordings: Results from the Colive Voice Study
    Ayadi, Hanin
    Elbéji, Abir
    Despotovic, Vladimir
    Fagherazzi, Guy
    Digital Biomarkers, 2024, 8 (01) : 159 - 170
  • [8] Voice based Detection of type 2 Diabetes Mellitus
    Chitkara, Divya
    Sharma, R. K.
    PROCEEDINGS OF THE 2016 IEEE 2ND INTERNATIONAL CONFERENCE ON ADVANCES IN ELECTRICAL & ELECTRONICS, INFORMATION, COMMUNICATION & BIO INFORMATICS (IEEE AEEICB-2016), 2016, : 83 - 87
  • [9] Use of Voice-Based Conversational Artificial Intelligence for Basal Insulin Prescription Management Among Patients With Type 2 Diabetes A Randomized Clinical Trial
    Nayak, Ashwin
    Vakili, Sharif
    Nayak, Kristen
    Nikolov, Margaret
    Chiu, Michelle
    Sosseinheimer, Philip
    Talamantes, Sarah
    Testa, Stefano
    Palanisamy, Srikanth
    Giri, Vinay
    Schulman, Kevin
    JAMA NETWORK OPEN, 2023, 6 (12) : E2340232
  • [10] Mixed-methods study protocol to identify expectations of people with type 1 diabetes and their caregivers about voice-based digital health solutions to support the management of diabetes distress: the PsyVoice study
    Vergara, Franchesca Aguirre
    Fischer, Aurelie
    Seuring, Till
    de Beaufort, Carine
    Fagherazzi, Guy
    Aguayo, Gloria A.
    BMJ OPEN, 2023, 13 (09):