EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

Prediction of cognitive impairment through speech data analysis: A comparative evaluation of deep learning models

Minsoo Kim, Hyunjoo Choi, YongSoo Shim, Nayoung Ryoo, Ho Tae Jeong, Gihyun Yun, Hunboc Lee, SangYun Kim iD, Young Chul Youn

PLOS One · 21.05.2026 · 21 (5) · e0349412

DOI: 10.1371/journal.pone.0349412 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift PLOS One, online erschienen am 21.05.2026. Verfasst von Minsoo Kim und 8 weiteren Autor:innen. Beteiligte Einrichtung: Research and Development, Baikal AI Inc., Seoul, Republic of Korea. Hauptthemen laut MEDLINE: Deep Learning, Cognitive Dysfunction, Speech, Alzheimer Disease. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of SMEs and Startups, Agriculture Science and Technology Development Rural Development Administration.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The early detection of cognitive impairments, such as mild cognitive impairment (MCI) and Alzheimer’s disease (AD), is essential for timely intervention and management. This study evaluates the performance of various deep-learning models in classifying speech recordings from individuals with normal cognition (NC), MCI, and AD, to identify the most effective approach for audio-based cognitive impairment diagnosis.

Methods

Speech data were obtained from the AI Hub “Cognitive Impairment Diagnosis Voice/Conversation” dataset. The study analyzed voice recordings from 320 female participants (105 with Alzheimer’s disease, 92 with mild cognitive impairment, and 123 cognitively normal controls). Three deep-learning architectures were compared: a one-dimensional convolutional neural network (1D CNN), an audio spectrogram transformer (AST), and a speech recognition model (Wav2Vec 2.0). The models were trained using features such as spectrograms, mel-spectrograms, and mel-frequency cepstral coefficients (MFCCs). Model performance was assessed using accuracy, recall, precision, and F1-score, with a five-fold cross-validation strategy to ensure robust and unbiased evaluation. Statistical significance was assessed using pairwise proportion z-tests with Holm-Bonferroni correction, and Wilson score 95% confidence intervals were computed for each model’s accuracy.

Results

Wav2Vec 2.0 outperformed the other models, achieving the highest accuracy and F1-scores for NC vs. MCI (accuracy: 0.74, F1: 0.72) and NC vs. AD (accuracy: 0.83, F1: 0.83). Pairwise proportion z-tests with Holm-Bonferroni correction confirmed that Wav2Vec 2.0 significantly outperformed 7 of 10 competing models (corrected p < 0.05) in both classification tasks. Performance varied by model and classification task, with Wav2Vec 2.0 consistently demonstrating superior accuracy across labels.

Conclusion

This study emphasizes the importance of selecting appropriate models and features for task-specific optimization and provides a foundation for developing non-invasive, speech-based diagnostic tools for cognitive disorders.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningCognitive DysfunctionSpeechAlzheimer DiseaseHumansFemaleConvolutional Neural NetworksAgedPredictive Learning ModelsAged, 80 and over

Autor:innen und Institutionen

  • Research and Development, Baikal AI Inc., Seoul, Republic of Korea.
  • Department of Communication Disorders, Korea Nazarene University, Cheonan, Republic of Korea.
  • Department of Neurology, The Catholic University of Korea, Seoul, Republic of Korea.
  • Department of Neurology, Chung-Ang University Hospital, Seoul, Republic of Korea.
  • Department of Neurology, Seoul National University College of Medicine and Seoul National University Bundang Hospital, Seongnam-si, Gyeonggi-do, Republic of Korea.
  • Department of Neurology, Chung-Ang University College of Medicine, Seoul, Republic of Korea.

ORCID-Profile: SangYun Kim

Förderung

  • Ministry of SMEs and Startups
  • Agriculture Science and Technology Development Rural Development Administration

Zitieren

APA

Kim, M., Choi, H., Shim, Y., Ryoo, N., Jeong, H. T., Yun, G., Lee, H., Kim, S., & Youn, Y. C. (2026). Prediction of cognitive impairment through speech data analysis: A comparative evaluation of deep learning models. PLOS One, 21(5), e0349412. https://doi.org/10.1371/journal.pone.0349412

Vancouver

Kim M, Choi H, Shim Y, Ryoo N, Jeong HT, Yun G, et al. Prediction of cognitive impairment through speech data analysis: A comparative evaluation of deep learning models. PLOS One. 2026;21(5):e0349412. doi:10.1371/journal.pone.0349412

BibTeX
@article{Kim2026,
  author = {Kim, Minsoo and Choi, Hyunjoo and Shim, YongSoo and Ryoo, Nayoung and Jeong, Ho Tae and Yun, Gihyun and Lee, Hunboc and Kim, SangYun and Youn, Young Chul},
  title = {Prediction of cognitive impairment through speech data analysis: A comparative evaluation of deep learning models},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {5},
  pages = {e0349412},
  doi = {10.1371/journal.pone.0349412}
}
Kostenlos starten →