Developing and testing AI-based voice biomarker models to detect cognitive impairment among community dwelling adults: a cross-sectional study in Japan.
| Title: | Developing and testing AI-based voice biomarker models to detect cognitive impairment among community dwelling adults: a cross-sectional study in Japan. |
|---|---|
| Authors: | Kiyoshige E; Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Ogata S; Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Kwon N; Canary Speech, Inc., 1800 Novell Place, Suite H51, Provo, UT, 84606, USA.; Nakaoku Y; Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Hayashi C; Research Institute of Nursing Care for People and Community, University of Hyogo, 13-71, Kitaoji-cho, Akashi, Hyogo, 673-8588, Japan.; Blaylock N; Canary Speech, Inc., 1800 Novell Place, Suite H51, Provo, UT, 84606, USA.; Brueckner R; Canary Speech, Inc., 1800 Novell Place, Suite H51, Provo, UT, 84606, USA.; Subramanian V; Canary Speech, Inc., 1800 Novell Place, Suite H51, Provo, UT, 84606, USA.; Joseph OConnell H; Canary Speech, Inc., 1800 Novell Place, Suite H51, Provo, UT, 84606, USA.; Yoshikawa Y; Department of Biostatistics, National Cerebral and Cardiovascular Centre, Suita, Osaka, 564-8565, Japan.; Teramoto K; Department of Biostatistics, National Cerebral and Cardiovascular Centre, Suita, Osaka, 564-8565, Japan.; Nakatsuka K; Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Saito S; Department of Neurology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Ihara M; Department of Neurology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan.; Takegami M; Department of Public Health and Health Policy, School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan.; Nishimura K; Department of Preventive Medicine and Epidemiology, National Cerebral and Cardiovascular Centre, 6-1 Kishibe-Simmachi, Suita, Osaka, 564-8565, Japan. |
| Source: | The Lancet regional health. Western Pacific [Lancet Reg Health West Pac] 2025 Jun 12; Vol. 59, pp. 101598. Date of Electronic Publication: 2025 Jun 12 (Print Publication: 2025). |
| Publication Type: | Journal Article |
| Language: | English |
| Journal Info: | Publisher: Elsevier Ltd Country of Publication: England NLM ID: 101774968 Publication Model: eCollection Cited Medium: Internet ISSN: 2666-6065 (Electronic) Linking ISSN: 26666065 NLM ISO Abbreviation: Lancet Reg Health West Pac Subsets: PubMed not MEDLINE |
| Imprint Name(s): | Original Publication: [London] : Elsevier Ltd., [2020]- |
| Abstract: | Background: Voice is a potential biomarker of cognitive impairment because mild cognitive impairment (MCI) can cause changes in speech patterns and tempo. Artificial intelligence (AI) can deliver voice biomarkers as prediction features, leading to a timely, noninvasive, and cost-effective detection of cognitive impairment. This study aimed to develop and test prediction models utilizing voice biomarkers to detect cognitive impairment, which AI derived from voice data of unstructured conversations in community-dwelling adults in Japan.; Methods: This observational study with a cross-sectional design, included 1461 community-dwelling adults. The outcome was cognitive impairment assessed by the Memory Performance Index score from the MCI screen. Voice data was collected from 3-min open-question interviews and extracted voice biomarkers based on acoustic and prosodic features as a 512-dimensional vector of individual voice information using the voice generator, Wav2Vec2. Other considerable predictors were age, sex, and education. We developed cognitive impairment prediction models by applying the extreme gradient boosting decision tree algorithm and a deep neural network model using 979 participants. Prediction performances were tested by area under the curves (AUCs) in 482 participants who were not used for model development.; Findings: We had 967 women (66·2%), 526 cognitive impairment (36·0%) participants with mean (standard deviation) age and education years of 79·5 (6·3) years old and 11·6 (2·2) years, respectively. The inclusion of voice biomarkers significantly improved AUCs (95% confidence intervals), from 0·80 (0·76, 0·84) to 0·88 (0·84, 0·91) for the age sex model and from 0·78 (0·73, 0·82) to 0·89 (0·86, 0·92) for the age sex and education model (p < 0·0001 for both comparisons by DeLong test).; Interpretation: Our prediction models for cognitive impairment using voice biomarkers can provide significantly timesaving MCI screening with high prediction performances (AUC = 0·89). Voice biomarkers significantly contributed to improving prediction performance.; Funding: Small Business Innovation Research (SBIR Phase 3 Fund), the Intramural Research Fund of Cardiovascular Diseases of the National Cerebral and Cardiovascular Center, and JSPS KAKENHI.; (© 2025 The Author(s).) |
| Competing Interests: | We have no conflicts of interest to declare. |
| References: | J Med Syst. 2023 Feb 1;47(1):17. (PMID: 36720727); Alzheimers Res Ther. 2023 Jul 22;15(1):128. (PMID: 37481563); Front Comput Sci. 2021;3:. (PMID: 35291512); Ann Geriatr Med Res. 2020 Sep;24(3):174-180. (PMID: 32829572); Lancet Healthy Longev. 2021 Jul;2(7):e407-e416. (PMID: 34240063); Ann Intern Med. 2015 Jan 6;162(1):W1-73. (PMID: 25560730); J Prim Care Community Health. 2022 Jan-Dec;13:21501319221117793. (PMID: 35950638); J Am Geriatr Soc. 2005 Apr;53(4):695-9. (PMID: 15817019); Int Psychogeriatr. 2019 Apr;31(4):491-504. (PMID: 30426911); Curr Alzheimer Res. 2020;17(1):60-68. (PMID: 32053074); PeerJ. 2021 Dec 20;9:e12656. (PMID: 35036144); Curr Alzheimer Res. 2018;15(2):120-129. (PMID: 28847279); Front Neurol. 2018 Nov 15;9:975. (PMID: 30498472); BMJ. 2024 Jan 22;384:e074821. (PMID: 38253388); Mayo Clin Proc. 2023 Sep;98(9):1353-1375. (PMID: 37661144); Curr Alzheimer Res. 2018;15(2):130-138. (PMID: 29165085); Dement Geriatr Cogn Disord. 2018;45(3-4):198-209. (PMID: 29886493); J Thorac Oncol. 2010 Sep;5(9):1315-6. (PMID: 20736804); Cochrane Database Syst Rev. 2015 Mar 05;(3):CD010783. (PMID: 25740785); Ann Palliat Med. 2021 Sep;10(9):9715-9724. (PMID: 34628897); PLoS One. 2021 Jul 14;16(7):e0253988. (PMID: 34260593); Lancet. 2024 Aug 10;404(10452):572-628. (PMID: 39096926); J Prev Alzheimers Dis. 2024;11(1):7-12. (PMID: 38230712); Alzheimers Dement. 2023 Jul;19(7):3235-3243. (PMID: 36934438); Cochrane Database Syst Rev. 2021 Jul 27;7:CD010783. (PMID: 34313331); J Neurol Neurosurg Psychiatry. 2019 Apr;90(4):373-379. (PMID: 29954871); Ann Intern Med. 2015 Jan 6;162(1):55-63. (PMID: 25560714); Alzheimers Dement. 2023 Mar;19(3):946-955. (PMID: 35796399); Proc Natl Acad Sci U S A. 2005 Mar 29;102(13):4919-24. (PMID: 15781874); Digit Biomark. 2020 Oct 19;4(3):99-108. (PMID: 33251474); Am J Alzheimers Dis Other Demen. 2008 Apr-May;23(2):162-6. (PMID: 18223126); Alzheimers Dement. 2009 Jul;5(4):295-306. (PMID: 19560100) |
| Contributed Indexing: | Keywords: AI; Mild cognitive impairment; Prediction model; Voice biomarkers |
| Entry Date(s): | Date Created: 20250717 Latest Revision: 20250719 |
| Update Code: | 20260130 |
| PubMed Central ID: | PMC12266181 |
| DOI: | 10.1016/j.lanwpc.2025.101598 |
| PMID: | 40673165 |
| Database: | MEDLINE |
Journal Article