{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:41Z","timestamp":1785545021053,"version":"3.56.0"},"reference-count":52,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Parkinson disease frequently manifests early vocal impairment, motivating the development of noninvasive and scalable digital screening tools.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study proposes a multiview spectrogram-based deep learning framework integrating recognition-aware context for Parkinson disease detection from voice recordings.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Voice recordings from 203 participants (121 with Parkinson disease and 82 healthy controls) were collected prospectively. Three spectrogram representations (Mel, short-time Fourier transform, and constant-Q transform) were extracted and processed through parallel convolutional neural network branches. A recognition ratio (RR) feature vector derived from automatic speech recognition transcript agreement was optionally fused with spectrogram embeddings. Models were evaluated using strict subject-wise 5-fold cross-validation.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n Multiview spectrogram recognition-aware Parkinson detection network achieved a mean test accuracy of 86.9% (SD 25.2%) using 3-view spectrogram fusion, improving to 97.4% (SD 5.7%) when incorporating the RR feature. RR integration reduced the false negative rate by approximately 84.5%, substantially improving sensitivity in screening-oriented settings.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Combining multiview spectrogram learning with recognition-aware context significantly enhances voice-based Parkinson disease classification under leakage-free evaluation. These findings support the potential of this approach for noninvasive screening in structured recording settings, while further validation in diverse real-world environments is needed.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/94063","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T15:22:10Z","timestamp":1781191330000},"page":"e94063-e94063","source":"Crossref","is-referenced-by-count":0,"title":["A Machine Learning Approach to Voice-Based Parkinson Disease Screening Using Multiview Spectrogram and Speech Recognition Features: Diagnostic Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-0253-1134","authenticated-orcid":false,"given":"Arifa","family":"Zahir","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2921-7233","authenticated-orcid":false,"given":"Jaehong","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9879-0634","authenticated-orcid":false,"given":"Jin-Sun","family":"Jun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3188-000X","authenticated-orcid":false,"given":"Kiwon","family":"Park","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8754-9180","authenticated-orcid":false,"given":"Ryul","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5641-8620","authenticated-orcid":false,"given":"Hyundoo","family":"Jeong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,6,11]]},"reference":[{"issue":"4","key":"R1","doi-asserted-by":"publisher","first-page":"368","DOI":"10.1136\/jnnp.2007.131045","article-title":"Parkinson\u2019s disease: clinical features and diagnosis","volume":"79","author":"Jankovic","journal-title":"J Neurol Neurosurg Psychiatry"},{"issue":"4","key":"R2","doi-asserted-by":"publisher","first-page":"884","DOI":"10.1109\/TBME.2009.2036000","article-title":"Accurate telemonitoring of Parkinson\u2019s disease progression by noninvasive speech tests","volume":"57","author":"Tsanas","journal-title":"IEEE Trans Biomed Eng"},{"issue":"4","key":"R3","doi-asserted-by":"publisher","first-page":"828","DOI":"10.1109\/JBHI.2013.2245674","article-title":"Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings","volume":"17","author":"Sakar","journal-title":"IEEE J Biomed Health Inform"},{"issue":"7","key":"R4","doi-asserted-by":"publisher","first-page":"1657","DOI":"10.1007\/s00521-015-2142-2","article-title":"A novel diagnosis system for Parkinson\u2019s disease using complex-valued artificial neural network with k-means clustering feature weighting method","volume":"28","author":"G\u00fcr\u00fcler","journal-title":"Neural Comput Applic"},{"issue":"5","key":"R5","doi-asserted-by":"publisher","DOI":"10.1007\/s10916-016-0477-6","article-title":"A decision support system to improve medical diagnosis using a combination of k-medoids clustering based attribute weighting and SVM","volume":"40","author":"Peker","journal-title":"J Med Syst"},{"issue":"8","key":"R6","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0182428","article-title":"Analyzing the effectiveness of vocal features in early telediagnosis of Parkinson\u2019s disease","volume":"12","author":"Erdogdu Sakar","journal-title":"PLoS One"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jocn.2019.12.029","article-title":"Voice changes in Parkinson\u2019s disease: what are they telling us?","volume":"72","author":"Ma","journal-title":"J Clin Neurosci"},{"issue":"1","key":"R8","doi-asserted-by":"publisher","first-page":"135","DOI":"10.3233\/JPD-230117","article-title":"Delivering multidisciplinary rehabilitation care in Parkinson\u2019s disease: an international consensus statement","volume":"14","author":"Goldman","journal-title":"J Parkinsons Dis"},{"issue":"3","key":"R9","doi-asserted-by":"publisher","first-page":"530","DOI":"10.1002\/ana.26835","article-title":"Spoken language alterations can predict phenoconversion in isolated rapid eye movement sleep behavior disorder: a multicenter study","volume":"95","author":"\u0160ubert","journal-title":"Ann Neurol"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.1016\/j.parkreldis.2022.01.003","article-title":"Voice characteristics from isolated rapid eye movement sleep behavior disorder to early Parkinson\u2019s disease","volume":"95","author":"Jeancolas","journal-title":"Parkinsonism Relat Disord"},{"issue":"2","key":"R11","doi-asserted-by":"publisher","DOI":"10.3390\/biomedicines10020371","article-title":"Emerging therapeutic strategies for Parkinson\u2019s disease and future prospects: a 2021 update","volume":"10","author":"Gouda","journal-title":"Biomedicines"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.bandl.2017.05.001","article-title":"Longitudinal decline in speech production in Parkinson\u2019s disease spectrum disorders","volume":"171","author":"Ash","journal-title":"Brain Lang"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","DOI":"10.1038\/s41531-022-00389-6","article-title":"Speech acoustic indices for differential diagnosis between Parkinson\u2019s disease, multiple system atrophy and progressive supranuclear palsy","volume":"8","author":"Daoudi","journal-title":"NPJ Parkinsons Dis"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.parkreldis.2023.105835","article-title":"Utilizing speech analysis to differentiate progressive supranuclear palsy from Parkinson\u2019s disease","volume":"115","author":"Kang","journal-title":"Parkinsonism Relat Disord"},{"issue":"8","key":"R15","doi-asserted-by":"publisher","first-page":"2101","DOI":"10.1111\/cns.14158","article-title":"Speech and gait abnormalities in motor subtypes of de-novo Parkinson\u2019s disease","volume":"29","author":"Rusz","journal-title":"CNS Neurosci Ther"},{"key":"R16","doi-asserted-by":"crossref","unstructured":"Ali MH Mohammed SL Al-Naji A . Parkinson\u2019s disease detection from voice using artificial intelligence techniques: a review. Presented at: The Fifth Scientific Conference for Electrical Engineering Techniques Research (EETR2024); Jun 15-16, 2024; Baghdad, Iraq. [doi: 10.1063\/5.0236188]","DOI":"10.1063\/5.0236188"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.3389\/fnagi.2025.1586273","article-title":"Non-invasive detection of Parkinson\u2019s disease based on speech analysis and interpretable machine learning","volume":"17","author":"Xu","journal-title":"Front Aging Neurosci"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.109566","article-title":"Multi-label speech feature selection for Parkinson\u2019s disease subtype recognition using graph model","volume":"185","author":"Ji","journal-title":"Comput Biol Med"},{"issue":"7","key":"R19","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering12070728","article-title":"Speech-based Parkinson\u2019s detection using pre-trained self-supervised automatic speech recognition (ASR) models and supervised contrastive learning","volume":"12","author":"Sedigh Malekroodi","journal-title":"Bioengineering (Basel)"},{"key":"R20","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2023.106700","article-title":"Detection of Parkinson\u2019s disease based on spectrograms of voice recordings and extreme learning machine random weight neural networks","volume":"125","author":"Guatelli","journal-title":"Eng Appl Artif Intell"},{"issue":"14","key":"R21","doi-asserted-by":"publisher","DOI":"10.3390\/s24144625","article-title":"Exploring spectrogram-based audio classification for Parkinson\u2019s disease: a study on speech classification and qualitative reliability verification","volume":"24","author":"Jeong","journal-title":"Sensors (Basel)"},{"issue":"10","key":"R22","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering12101052","article-title":"Voice-based early diagnosis of Parkinson\u2019s disease using spectrogram features and AI models","volume":"12","author":"Quamar","journal-title":"Bioengineering (Basel)"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-17863-z","article-title":"Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks","volume":"12","author":"Qu","journal-title":"Sci Rep"},{"issue":"4","key":"R24","doi-asserted-by":"publisher","DOI":"10.1007\/s40747-025-01800-4","article-title":"CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning","volume":"11","author":"Chu","journal-title":"Complex Intell Syst"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Zhao C Ma S Li D et al. 3D Mel-spectrogram\u2013based deep learning for automated multiclass diagnosis of pathological voices. Research Square. Preprint posted online on Oct 28, 2025. [doi: 10.21203\/rs.3.rs-7711140\/v1]","DOI":"10.21203\/rs.3.rs-7711140\/v1"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2025.1706317","article-title":"Parkinson\u2019s disease detection using spectrogram-based multi-model feature fusion networks","volume":"16","author":"Chen","journal-title":"Front Neurol"},{"issue":"2","key":"R27","doi-asserted-by":"publisher","first-page":"135","DOI":"10.3390\/info16020135","article-title":"PD-Net: Parkinson\u2019s disease detection through fusion of two spectral features using attention-based hybrid deep neural network","volume":"16","author":"Islam","journal-title":"Information"},{"issue":"11","key":"R28","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering12111279","article-title":"Voice-based detection of Parkinson\u2019s disease using machine and deep learning approaches: a systematic review","volume":"12","author":"Sedigh Malekroodi","journal-title":"Bioengineering (Basel)"},{"key":"R29","doi-asserted-by":"publisher","DOI":"10.2196\/71560","article-title":"Digital biomarkers for Parkinson disease: bibliometric analysis and a scoping review of deep learning for freezing of gait","volume":"27","author":"Qi","journal-title":"J Med Internet Res"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-21407-6","article-title":"Multi-modal deep learning framework for early detection of Parkinson\u2019s disease using neurological and physiological data for high-fidelity diagnosis","volume":"15","author":"Sar","journal-title":"Sci Rep"},{"issue":"1","key":"R31","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01217-2","article-title":"Digital biomarkers for precision diagnosis and monitoring in Parkinson\u2019s disease","volume":"7","author":"Sun","journal-title":"NPJ Digit Med"},{"key":"R32","doi-asserted-by":"crossref","unstructured":"Sushmitha Saro R Jaya Suriya B Rajakumari R . Comprehensive speech emotion recognition system employing multi-layer perceptron (MLP) classifier and librosa feature extraction. Presented at: 2023 International Conference on Sustainable Communication Networks and Application (ICSCNA); Nov 15-17, 2023; Theni, India. [doi: 10.1109\/ICSCNA58489.2023.10370394]","DOI":"10.1109\/ICSCNA58489.2023.10370394"},{"issue":"3","key":"R33","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1155\/1999\/327643","volume":"11","author":"Ho","journal-title":"Behav Neurol"},{"issue":"3","key":"R34","doi-asserted-by":"publisher","first-page":"279","DOI":"10.2217\/nmt.12.15","article-title":"Speech, voice and language in Parkinson\u2019s disease: changes and interventions","volume":"2","author":"Miller","journal-title":"Neurodegen Dis Manage"},{"issue":"2","key":"R35","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1007\/s00702-009-0351-5","article-title":"Short- and long-term dopaminergic effects on dysarthria in early Parkinson\u2019s disease","volume":"117","author":"Skodda","journal-title":"J Neural Transm (Vienna)"},{"issue":"6","key":"R36","doi-asserted-by":"publisher","first-page":"427","DOI":"10.1080\/0269920031000086248","article-title":"Toward an acoustic typology of motor speech disorders","volume":"17","author":"Kent","journal-title":"Clin Linguist Phon"},{"key":"R37","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2021.103006","article-title":"Parkinson\u2019s detection based on combined CNN and LSTM using enhanced speech signals with variational mode decomposition","volume":"70","author":"Er","journal-title":"Biomed Signal Process Control"},{"issue":"5","key":"R38","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering10050531","article-title":"CNN-based identification of Parkinson\u2019s disease from continuous speech in noisy environments","volume":"10","author":"Farag\u00f3","journal-title":"Bioengineering (Basel)"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-92105-6","article-title":"Pre-trained convolutional neural networks identify Parkinson\u2019s disease from spectrogram images of voice samples","volume":"15","author":"Rahmatallah","journal-title":"Sci Rep"},{"issue":"2","key":"R40","doi-asserted-by":"publisher","first-page":"163","DOI":"10.14744\/SEMB.2023.29560","article-title":"Hypokinetic dysarthria in Parkinson\u2019s disease: a narrative review","volume":"57","author":"Atalar","journal-title":"Sisli Etfal Hastan Tip Bul"},{"issue":"1","key":"R41","doi-asserted-by":"publisher","DOI":"10.1038\/s41531-025-00913-4","article-title":"Speech and language biomarkers for Parkinson\u2019s disease prediction, early diagnosis and progression","volume":"11","author":"Cao","journal-title":"NPJ Parkinsons Dis"},{"issue":"23","key":"R42","doi-asserted-by":"publisher","first-page":"11085","DOI":"10.3390\/app142311085","article-title":"Speech signal analysis in patients with Parkinson\u2019s disease, taking into account phonation, articulation, and prosody of speech","volume":"14","author":"Majda-Zdancewicz","journal-title":"Appl Sci"},{"key":"R43","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2023.1129830","article-title":"Vowel production: a potential speech biomarker for early detection of dysarthria in Parkinson\u2019s disease","volume":"14","author":"Roland","journal-title":"Front Psychol"},{"issue":"5","key":"R44","doi-asserted-by":"publisher","first-page":"1481","DOI":"10.1111\/ene.15743","article-title":"A systematic review and meta-analysis of respiratory dysfunction in Parkinson\u2019s disease","volume":"30","author":"McMahon","journal-title":"Eur J Neurol"},{"issue":"4","key":"R45","doi-asserted-by":"publisher","DOI":"10.1007\/s00415-025-13008-0","article-title":"Dyspnea and dystussia in Parkinson\u2019s disease: patient-reported prevalence and determinants","volume":"272","author":"van de Wetering-van Dongen","journal-title":"J Neurol"},{"key":"R46","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2025.108954","article-title":"Voice classification in Parkinson\u2019s disease: a deep learning approach using transformers and error rate metrics","volume":"113","author":"Perrone","journal-title":"Biomed Signal Process Control"},{"issue":"13","key":"R47","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13132163","article-title":"Deep learning and artificial intelligence applied to model speech and language in Parkinson\u2019s disease","volume":"13","author":"Escobar-Grisales","journal-title":"Diagnostics (Basel)"},{"issue":"10","key":"R48","doi-asserted-by":"publisher","DOI":"10.3390\/jimaging11100341","article-title":"Multi-channel spectro-temporal representations for speech-based Parkinson\u2019s disease detection","volume":"11","author":"Malekroodi","journal-title":"J Imaging"},{"key":"R49","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2021.115013","article-title":"Robust automated Parkinson disease detection based on voice signals with transfer learning","volume":"178","author":"Karaman","journal-title":"Expert Syst Appl"},{"issue":"3","key":"R50","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering11030295","article-title":"Leveraging deep learning for fine-grained categorization of Parkinson\u2019s disease progression levels through analysis of vocal acoustic patterns","volume":"11","author":"Malekroodi","journal-title":"Bioengineering (Basel)"},{"key":"R51","doi-asserted-by":"publisher","DOI":"10.1016\/j.apacoust.2023.109657","article-title":"High-resolution superlet transform based techniques for Parkinson\u2019s disease detection using speech signal","volume":"214","author":"Bhatt","journal-title":"Appl Acoust"},{"issue":"3","key":"R52","doi-asserted-by":"publisher","first-page":"657","DOI":"10.1007\/s10772-024-10128-2","article-title":"A hybrid approach to detecting Parkinson\u2019s disease using spectrogram and deep learning CNN-LSTM network","volume":"27","author":"Shibina","journal-title":"Int J Speech Technol"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e94063","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:23:34Z","timestamp":1785536614000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e94063"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,11]]},"references-count":52,"URL":"https:\/\/doi.org\/10.2196\/94063","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,11]]},"article-number":"v14i4e94063"}}