{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T13:11:58Z","timestamp":1784207518059,"version":"3.55.0"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T00:00:00Z","timestamp":1784160000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurol."],"abstract":"<jats:p>Neurological disorders can alter motor control, cognition, and speech production in subtle ways that are difficult to detect using standard diagnostic tools. Because conventional imaging techniques such as CT and MRI often fail to reveal functional abnormalities, there is growing interest in non-invasive digital health approaches. This study investigates the potential of speech-based deep learning for detecting concussion-related dysfunction using structured speech tasks. Audio recordings from 225 high school and college athletes (approximately 200 retained after quality filtering) performing eight speech tasks were analyzed. The recordings were converted into Mel spectrograms and processed using a convolutional neural network (CNN), with data augmentation applied to improve model generalization. The results demonstrate strong classification performance, particularly for tasks requiring dynamic articulation and rhythmic control. To better interpret model behavior, the speech tasks were grouped into three functional categories and examined using gradient-weighted class activation mapping (Grad-CAM). Sentence and word reading tasks, as well as rapid syllable repetition, achieved the highest accuracy (up to 96%) and exhibited focused activation patterns associated with phonetic complexity and motor timing. In contrast, sustained vowel tasks produced lower performance and more diffuse model attention, likely due to their limited acoustic variability. These findings highlight the effectiveness of CNN-based analysis of dynamic speech features and emphasize the importance of task selection in speech-based digital diagnostics. Although the framework may inform future speech-based neurological assessment research, this study focuses on mild traumatic brain injury (concussion) as an initial use case.<\/jats:p>","DOI":"10.3389\/fneur.2026.1830075","type":"journal-article","created":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T12:12:51Z","timestamp":1784203971000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Speech-based concussion detection in athletes using Mel-spectrograms and convolutional neural networks"],"prefix":"10.3389","volume":"17","author":[{"given":"Rahmina","family":"Rubaiat","sequence":"first","affiliation":[{"name":"Mobile Sensing and Analytics Lab, Knight Foundation School of Computing and Information Sciences, Florida International University","place":["Miami, FL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christian","family":"Poellabauer","sequence":"additional","affiliation":[{"name":"Mobile Sensing and Analytics Lab, Knight Foundation School of Computing and Information Sciences, Florida International University","place":["Miami, FL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,16]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"695","DOI":"10.1136\/bjsports-2023-106898","article-title":"Consensus statement on concussion in sport: the 6th International Conference on Concussion in Sport\u2013Amsterdam, October 2022","volume":"57","author":"Patricios","year":"2023","journal-title":"Br J Sports Med."},{"key":"B2","doi-asserted-by":"publisher","first-page":"S50","DOI":"10.1227\/NEU.0000000000000491","article-title":"Imaging concussion: a review","volume":"75","author":"Yuh","year":"2014","journal-title":"Neurosurgery."},{"key":"B3","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1007\/s40120-019-00153-8","article-title":"Deep learning and neurology: a systematic review","volume":"8","author":"Valliani","year":"2019","journal-title":"Neurol Ther."},{"key":"B4","doi-asserted-by":"publisher","first-page":"842301","DOI":"10.3389\/fdgth.2022.842301","article-title":"Voice analysis for neurological disorder recognition-a systematic review and perspective on emerging trends","volume":"4","author":"Pascal","year":"2022","journal-title":"Front. Digit. Health."},{"key":"B5","doi-asserted-by":"publisher","first-page":"620251","DOI":"10.3389\/fpsyg.2021.620251","article-title":"Ten years of research on automatic voice and speech analysis of people with Alzheimer's disease and mild cognitive impairment: a systematic review article","volume":"12","author":"Mart\u00ednez-Nicol\u00e1s","year":"2021","journal-title":"Front Psychol."},{"key":"B6","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1159\/000510820","article-title":"Evaluation of speech-based digital biomarkers: review and recommendations","volume":"4","author":"Robin","year":"2020","journal-title":"Digit Biomark."},{"key":"B7","doi-asserted-by":"publisher","first-page":"700","DOI":"10.1109\/JSTSP.2025.3591062","article-title":"Overview of automatic speech analysis and technologies for neurodegenerative disorders: diagnosis and assistive applications","volume":"19","author":"Sheikh","year":"2025"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1449388","DOI":"10.3389\/fnhum.2024.1449388","article-title":"D\u00f6rr F, Garthof S, Sch\u00e4fer S, Elmers J, Schwed L, et al. Detecting fatigue in multiple sclerosis through automatic speech analysis","volume":"18","author":"Dias","year":"2024","journal-title":"Front Hum Neurosci."},{"key":"B9","doi-asserted-by":"publisher","first-page":"325","DOI":"10.3766\/jaaa.16051","article-title":"Auditory and cognitive factors associated with speech-in-noise complaints following mild traumatic brain injury","volume":"28","author":"Hoover","year":"2017","journal-title":"J Am Acad Audiol."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1101\/2020.06.14.20130443","article-title":"Sports related concussion impacts speech rate and muscle function","author":"Banks","year":"2020","journal-title":"medRxiv."},{"key":"B11","doi-asserted-by":"publisher","first-page":"496","DOI":"10.1109\/JBHI.2016.2633509","article-title":"Portable mTBI assessment using temporal and frequency analysis of speech","volume":"21","author":"Daudet","year":"2016","journal-title":"IEEE J Biomed Health Inform."},{"key":"B12","doi-asserted-by":"publisher","first-page":"1135441","DOI":"10.3389\/fpsyg.2023.1135441","article-title":"Speech error rates after a sports-related concussion","volume":"14","author":"Patel","year":"2023","journal-title":"Front Psychol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1371\/journal.pone.0274395","article-title":"A deep learning-based approach to diagnose mild traumatic brain injury using audio classification","volume":"17","author":"Wall","year":"2022","journal-title":"PLoS ONE."},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1101\/2021.09.20.21263753","article-title":"Detecting a history of repetitive head impacts from a short voice recording","author":"Tauro","year":"2021","journal-title":"medRxiv."},{"key":"B15","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med Image Anal."},{"key":"B16","doi-asserted-by":"publisher","first-page":"5582","DOI":"10.1109\/ICASSP.2014.6854671","article-title":"Data augmentation for deep neural network acoustic modeling","volume":"23","author":"Cui","year":"2014","journal-title":"IEEE\/ACM Trans Audio Speech Lang Process."},{"key":"B17","article-title":"Audio augmentation for speech recognition","author":"Tom","year":"2015","journal-title":"INTERSPEECH 2015; Dresden"},{"key":"B18","doi-asserted-by":"publisher","first-page":"726","DOI":"10.14569\/IJACSA.2020.0111089","article-title":"Deep acoustic embeddings for identifying parkinsonian speech","volume":"11","author":"Syed","year":"2020","journal-title":"Int J Adv Comput Sci Appl."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1351","article-title":"A deep learning method for pathological voice detection using convolutional deep belief networks","author":"Wu","year":"2018","journal-title":"INTERSPEECH 2018; Hyderabad"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1319","DOI":"10.1016\/j.aej.2021.06.024","article-title":"Automatic diagnosis of COVID-19 disease using deep convolutional neural network with multi-feature channel from respiratory sound data: cough, voice, and breath","volume":"61","author":"Lella","year":"2022","journal-title":"Alex Eng J."},{"key":"B21","doi-asserted-by":"publisher","first-page":"105550","DOI":"10.1016\/j.compbiomed.2022.105550","article-title":"Explainable detection of myocardial infarction using deep learning models with Grad-CAM technique on ECG signals","volume":"146","author":"Jahmunah","year":"2022","journal-title":"Comput Biol Med."},{"key":"B22","volume-title":"Accurate Telemonitoring of Parkinson's Disease Symptom Severity Using Nonlinear Speech Signal Processing and Statistical Machine Learning","author":"Tsanas","year":"2012"},{"key":"B23","doi-asserted-by":"publisher","first-page":"130","DOI":"10.2174\/1567205014666171121114930","article-title":"A speech recognition-based solution for the automatic detection of mild cognitive impairment from spontaneous speech","volume":"15","author":"Toth","year":"2018","journal-title":"Curr Alzheimer Res."},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680","article-title":"Specaugment: a simple data augmentation method for automatic speech recognition","author":"Park","year":"2019","journal-title":"arXiv"},{"key":"B25","author":"Eklund","year":"2019","journal-title":"Data Augmentation Techniques for Robust Audio Analysis."},{"key":"B26","doi-asserted-by":"publisher","first-page":"214","DOI":"10.21849\/cacd.2019.00150","article-title":"Motor speech and motor limb status in athletes following a concussion","volume":"4","author":"Salvatore","year":"2019","journal-title":"Clin Arch Commun Disord."},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-789","article-title":"A portable automatic PA-TA-KA syllable detection system to derive biomarkers for neurological disorders","author":"Tao","year":"2016","journal-title":"INTERSPEECH 2016; San Francisco, CA"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1581","article-title":"Acoustic indicators of speech motor coordination in adults with and without traumatic brain injury","author":"Talkar1","year":"2021","journal-title":"INTERSPEECH 2021; Brno"},{"key":"B29","doi-asserted-by":"publisher","first-page":"994","DOI":"10.1044\/2016_JSLHR-S-15-0307","article-title":"Comparing measures of voice quality from sustained phonation and continuous speech","volume":"59","author":"Gerratt","year":"2016","journal-title":"J Speech Lang Hear Res."},{"key":"B30","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1016\/j.jvoice.2011.11.001","article-title":"Perception of vocal tremor during sustained phonation compared with sentence context","volume":"26","author":"Lederle","year":"2012","journal-title":"J Voice."},{"key":"B31","doi-asserted-by":"publisher","first-page":"1143","DOI":"10.1109\/ACCESS.2015.2457392","article-title":"Challenges in concussion detection using vocal acoustic biomarkers","volume":"3","author":"Poellabauer","year":"2015","journal-title":"IEEE Access."},{"key":"B32","doi-asserted-by":"publisher","first-page":"734501","DOI":"10.3389\/fnhum.2021.734501","article-title":"Deep learning recurrent neural network for concussion classification in adolescents using raw electroencephalography signals: Toward a minimal number of sensors","volume":"15","author":"Thanjavur","year":"2021","journal-title":"Front. Hum. Neurosci."}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1830075\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T12:12:53Z","timestamp":1784203973000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1830075\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,16]]},"references-count":32,"alternative-id":["10.3389\/fneur.2026.1830075"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2026.1830075","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,16]]},"article-number":"1830075"}}