{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T21:18:20Z","timestamp":1793395100640,"version":"4.5.0"},"reference-count":42,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2024,2,25]],"date-time":"2024-02-25T00:00:00Z","timestamp":1708819200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Institute on Aging","award":["NIA AG03705"],"award-info":[{"award-number":["NIA AG03705"]}]},{"name":"National Institute on Aging","award":["AG05133"],"award-info":[{"award-number":["AG05133"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Brain Sciences"],"abstract":"<jats:p>The development of noninvasive and cost-effective methods of detecting Alzheimer\u2019s disease (AD) is essential for its early prevention and mitigation. We optimize the detection of AD using natural language processing (NLP) of spontaneous speech through the use of audio enhancement techniques and novel transcription methodologies. Specifically, we utilized Boll Spectral Subtraction to improve audio fidelity and created transcriptions using state-of-the-art AI services\u2014locally-based Wav2Vec and Whisper, alongside cloud-based IBM Cloud and Rev AI\u2014evaluating their performance against traditional manual transcription methods. Support Vector Machine (SVM) classifiers were then trained and tested using GPT-based embeddings of transcriptions. Our findings revealed that AI-based transcriptions largely outperformed traditional manual ones, with Wav2Vec (enhanced audio) achieving the best accuracy and F-1 score (0.99 for both metrics) for locally-based systems and Rev AI (standard audio) performing the best for cloud-based systems (0.96 for both metrics). Furthermore, this study revealed the detrimental effects of interviewer speech on model performance in addition to the minimal effect of audio enhancement. Based on our findings, current AI transcription and NLP technologies are highly effective at accurately detecting AD with available data but struggle to classify probable AD and mild cognitive impairment (MCI), a prodromal stage of AD, due to a lack of training data, laying the groundwork for the future implementation of an automatic AD detection system.<\/jats:p>","DOI":"10.3390\/brainsci14030211","type":"journal-article","created":{"date-parts":[[2024,2,26]],"date-time":"2024-02-26T06:50:23Z","timestamp":1708930223000},"page":"211","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":22,"title":["The Optimization of a Natural Language Processing Approach for the Automatic Detection of Alzheimer\u2019s Disease Using GPT Embeddings"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-0313-5951","authenticated-orcid":false,"given":"Benjamin S.","family":"Runde","sequence":"first","affiliation":[{"name":"Science Engineering Research Center, The Potomac School, McLean, VA 22101, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ajit","family":"Alapati","sequence":"additional","affiliation":[{"name":"Neuroscience Center of Excellence, School of Medicine, New Orleans, LA 70112, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9243-5444","authenticated-orcid":false,"given":"Nicolas G.","family":"Bazan","sequence":"additional","affiliation":[{"name":"Neuroscience Center of Excellence, School of Medicine, New Orleans, LA 70112, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2024,2,25]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"(2023). 2023 Alzheimer\u2019s disease facts and figures. Alzheimer\u2019s Dement., 19, 1598\u20131695.","DOI":"10.1002\/alz.13016"},{"key":"ref_2","unstructured":"(2024, January 02). World Alzheimer Report 2023 Reducing Dementia Risk: Never too Early, Never too Late. Available online: https:\/\/www.alzint.org\/u\/World-Alzheimer-Report-2023.pdf."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"561","DOI":"10.1016\/j.jalz.2014.06.004","article-title":"Innovative diagnostic tools for early detection of Alzheimer\u2019s disease","volume":"11","author":"Laske","year":"2014","journal-title":"Alzheimer\u2019s Dement."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Vigo, I., Coelho, L., and Reis, S. (2022). Speech- and Language-Based Classification of Alzheimer\u2019s Disease: A Systematic Review. Bioengineering, 9.","DOI":"10.3390\/bioengineering9010027"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1590\/S1980-57642014DN83000010","article-title":"Analysis of word number and content in discourse of patients with mild to moderate Alzheimer\u2019s disease","volume":"8","author":"Onofre","year":"2014","journal-title":"Dement. Neuropsychol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1002\/pnp.721","article-title":"Artificial intelligence as an aid to diagnosing dementia: An overview","volume":"25","author":"Dashwood","year":"2021","journal-title":"Prog. Neurol. Psychiatry"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"544","DOI":"10.1136\/amiajnl-2011-000464","article-title":"Natural language processing: An introduction","volume":"18","author":"Nadkarni","year":"2011","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Jiao, Q., and Zhang, S. (2021, January 12\u201314). A Brief Survey of Word Embedding and Its Recent Development. Proceedings of the 2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China.","DOI":"10.1109\/IAEAC50856.2021.9390956"},{"key":"ref_9","unstructured":"Neelakantan, A., Xu, T., Puri, R., Radford, A., Han, J.M., Tworek, J., Yuan, Q., Tezak, N., Kim, J., and Hallacy, C. (2024, January 02). Text and Code Embeddings by Contrastive Pre-Training. Available online: https:\/\/arxiv.org\/pdf\/2201.10005.pdf."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Agbavor, F., and Liang, H. (2022). Predicting dementia from spontaneous speech using large language models. PLoS Digital Health, 1.","DOI":"10.1371\/journal.pdig.0000168"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Luz, S., Haider, F., de la Fuente Garcia, S., Fromm, D., and MacWhinney, B. (2021). Editorial: Alzheimer\u2019s Dementia Recognition through Spontaneous Speech. Front. Comput. Sci., 3.","DOI":"10.3389\/fcomp.2021.780169"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"623237","DOI":"10.3389\/fpsyg.2020.623237","article-title":"Towards Computer-Based Automated Screening of Dementia Through Spontaneous Speech","volume":"11","author":"Chlasta","year":"2021","journal-title":"Front. Psychol."},{"key":"ref_13","unstructured":"Balagopalan, A., and Novikova, J. (2024, January 02). Comparing Acoustic-based Approaches for Alzheimer\u2019s Disease Detection. Available online: https:\/\/arxiv.org\/pdf\/2106.01555.pdf."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Santander-Cruz, Y., Salazar-Colores, S., Paredes-Garc\u00eda, W.J., Guendulain-Arenas, H., and Tovar-Arriaga, S. (2022). Semantic Feature Extraction Using SBERT for Dementia Detection. Brain Sci., 12.","DOI":"10.3390\/brainsci12020270"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"585","DOI":"10.1001\/archneur.1994.00540180063015","article-title":"The natural history of Alzheimer\u2019s disease: Description of study cohort and accuracy of diagnosis","volume":"51","author":"Becker","year":"1994","journal-title":"Arch. Neurol."},{"key":"ref_16","unstructured":"(2024, January 02). DementiaBank: Theoretical Rationale, Protocol, and Illustrative Analyses. Available online: https:\/\/pubs.asha.org\/doi\/10.1044\/2022_AJSLP-22-00281."},{"key":"ref_17","unstructured":"(2024, January 02). Boll Spectral Subtraction. File Exchange-MATLAB Central File Exchange-MATLAB Central. Available online: https:\/\/www.mathworks.com\/matlabcentral\/fileexchange\/7675-boll-spectral-subtraction."},{"key":"ref_18","unstructured":"Goodglass, H., Kaplan, E., and Weintraub, S. (2001). BDAE: The Boston Diagnostic Aphasia Examination, Lippincott Williams & Wilkins."},{"key":"ref_19","unstructured":"Cummings, L. (2024, January 02). Describing the Cookie Theft picture: Sources of Breakdown in Alzheimer\u2019s Dementia. Available online: https:\/\/www.researchgate.net\/publication\/332061806_Describing_the_Cookie_Theft_picture_Sources_of_breakdown_in_Alzheimer\u2019s_dementia."},{"key":"ref_20","unstructured":"(2023, October 24). MATLAB Version: 9.13.0 (R2022b). Available online: https:\/\/www.mathworks.com."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1109\/TASSP.1979.1163209","article-title":"Suppression of acoustic noise in speech using spectral subtraction","volume":"27","author":"Boll","year":"1979","journal-title":"IEEE Trans. Acoust. Speech, Signal Process."},{"key":"ref_22","unstructured":"Budach, L., Feuerpfeil, M., Ihde, N., Nathansen, A., Noack, N., Patzlaff, H., Naumann, F., and Harmouch, H. (2024, January 02). The Effects of Data Quality on Machine Learning Performance. Available online: https:\/\/arxiv.org\/abs\/2207.14529."},{"key":"ref_23","unstructured":"Baevski, A., Zhou, H., Mohamed, A., and Auli, M. (2024, January 02). wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations. Available online: https:\/\/arxiv.org\/abs\/2006.11477."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., and Funtowicz, M. (, January October). Transformers: State-of-the-Art Natural Language Processing. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, Online.","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Mcfee, B., Raffel, C., Liang, D., Ellis, D., Mcvicar, M., Battenberg, E., and Nieto, O. (2015, January 6\u201312). librosa: Audio and Music Signal Analysis in Python. Proceedings of the 14th Python in Science Conference, Austin, TX, USA.","DOI":"10.25080\/Majora-7b98e3ed-003"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"426","DOI":"10.1044\/2022_AJSLP-22-00281","article-title":"DementiaBank: Theoretical Rationale, Protocol, and Illustrative Analyses","volume":"32","author":"Lanzi","year":"2023","journal-title":"Am. J. Speech-Lang. Pathol."},{"key":"ref_27","first-page":"2","article-title":"Docker: Lightweight Linux containers for consistent development and deployment","volume":"2014","author":"Merkel","year":"2014","journal-title":"Linux J."},{"key":"ref_28","unstructured":"(2024, January 02). Speech to Text API | Speech Recognition Service\u2014Rev AI. Available online: https:\/\/www.rev.ai\/."},{"key":"ref_29","unstructured":"(2024, January 02). Global Speech-to-Text Transcript Error Rating 2021|Statista. Available online: https:\/\/www.statista.com\/statistics\/1133833\/speech-to-text-transcript-accuracy-rate-among-leading-companies\/."},{"key":"ref_30","unstructured":"Radford, A., Kim, J., Xu, T., Brockman, G., McLeavey, C., and Sutskever, I. (2024, January 02). Robust Speech Recognition Via Large-Scale Weak Supervision. Available online: https:\/\/cdn.openai.com\/papers\/whisper.pdf."},{"key":"ref_31","unstructured":"Liddell, R. (2024, January 02). Next-Generation Watson Speech to Text\u2014IBM Watson Speech Services. Available online: https:\/\/medium.com\/ibm-watson-speech-services\/next-generation-watson-speech-to-text-650fd66d95d0."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"McKinney, W. (2010, January 28\u201330). Data structures for statistical computing in python. Proceedings of the 9th Python in Science Conference, Austin, TX, USA.","DOI":"10.25080\/Majora-92bf1922-00a"},{"key":"ref_33","unstructured":"(2024, January 02). OpenAI Platform. Available online: https:\/\/platform.openai.com\/docs\/guides\/embeddings\/what-are-embeddings."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic minority over-sampling technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intell. Res."},{"key":"ref_35","unstructured":"(2024, January 02). Imbalanced-Learn Documentation\u2014Version 0.11.0. Available online: https:\/\/imbalanced-learn.org\/stable\/."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Blagus, R., and Lusa, L. (2013). SMOTE for High-Dimensional Class-Imbalanced Data. BMC Bioinform., 14.","DOI":"10.1186\/1471-2105-14-106"},{"key":"ref_37","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1038\/s41586-020-2649-2","article-title":"Array programming with NumPy","volume":"585","author":"Harris","year":"2020","journal-title":"Nature"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"Matplotlib: A 2D Graphics Environment","volume":"9","author":"Hunter","year":"2007","journal-title":"Comput. Sci. Eng."},{"key":"ref_40","unstructured":"Raschka, S. (2024, January 02). An Overview of General Performance Metrics of Binary Classifier Systems. Available online: https:\/\/arxiv.org\/pdf\/1410.5330.pdf."},{"key":"ref_41","unstructured":"Luz, S., Haider, F., de la Fuente, S., Fromm, D., and MacWhinney, B. (2024, January 02). Detecting Cognitive Decline Using Speech Only: The ADReSSo Challenge. Available online: https:\/\/arxiv.org\/abs\/2104.09356."},{"key":"ref_42","unstructured":"Valdenegro-Toro, M., and Sabatelli, M. (2023, November 08). Machine Learning Students Overfit to Overfitting. Available online: https:\/\/arxiv.org\/pdf\/2209.03032.pdf."}],"container-title":["Brain Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2076-3425\/14\/3\/211\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T14:04:45Z","timestamp":1760105085000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2076-3425\/14\/3\/211"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,25]]},"references-count":42,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2024,3]]}},"alternative-id":["brainsci14030211"],"URL":"https:\/\/doi.org\/10.3390\/brainsci14030211","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2024.01.14.24301297","asserted-by":"object"}]},"ISSN":["2076-3425"],"issn-type":[{"value":"2076-3425","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,25]]}}}