{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,5]],"date-time":"2026-11-05T14:25:07Z","timestamp":1793888707034,"version":"4.8.0"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T00:00:00Z","timestamp":1788912000000},"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. Aging Neurosci."],"abstract":"<jats:p>\n                    Parkinson's disease (PD) is a common neuro-degenerative disorder. Recent studies have used non-invasive biomarkers such as electroencephalography (EEG) and speech signals for PD diagnosis. However, majority of them optimize models at the instance level, whereas clinical diagnosis is a subject-level decision making process. In this work, we propose MIL-O-PD, a novel two-stage subject-level multimodal framework that formulates PD diagnosis. The stage-1 implements a Multiple Instance Learning approach with modality specific encoders and an attention-based aggregation mechanism. Stage-2 focuses incorporating a Gray Wolf Optimization-based\n                    <jats:italic>post-hoc<\/jats:italic>\n                    feature optimization module and final subject-level classifier. It supports representation-level multimodal fusion under strict subject-independent conditions. The proposed framework achieves up to 70.5% subject-level accuracy and 69.7% F1-score, while outperforming alternate pooling mechanisms. Further, the analysis of attention weights highlights modality-specific behavior, supporting the interpretability of the model. Overall, this study presents a principled framework for subject-level analysis of PD designed for unpaired multimodal data addressing the importance of aligning learning objectives with real-world diagnostic processes.\n                  <\/jats:p>","DOI":"10.3389\/fnagi.2026.1924167","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T10:16:37Z","timestamp":1788948997000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MIL-O-PD: a two-stage multiple instance learning and heuristic optimization framework for unpaired multimodal Parkinson's diagnosis"],"prefix":"10.3389","volume":"18","author":[{"given":"Sankhadip","family":"Bera","sequence":"first","affiliation":[{"name":"Department of Information Technology, Jadavpur University, Salt Lake Campus","place":["Kolkata, West Bengal, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Fazal","family":"Ijaz","sequence":"additional","affiliation":[{"name":"School of Technology, Business and Hospitality Faculty, Torrens University, Campus Flinders, Melbourne","place":["VIC, Australia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaeyoung","family":"Choi","sequence":"additional","affiliation":[{"name":"School of Computing, Gachon University","place":["Seongnam-si, Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pawan Kumar","family":"Singh","sequence":"additional","affiliation":[{"name":"Department of Information Technology, Jadavpur University, Salt Lake Campus","place":["Kolkata, West Bengal, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,9]]},"reference":[{"key":"B1","unstructured":""},{"key":"B2","doi-asserted-by":"publisher","first-page":"21390","DOI":"10.1038\/s41598-025-06121-7","article-title":"Schizophrenia detection from electroencephalogram signals using image encoding and wrapper-based deep feature selection approach","volume":"15","author":"Aich","year":"2025","journal-title":"Sci. Rep"},{"key":"B3","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1038\/s41531-023-00602-0","article-title":"Resting-state EEG measures cognitive impairment in Parkinson's disease","volume":"10","author":"Anjum","year":"2024","journal-title":"NPJ Parkinsons Dis"},{"key":"B4","unstructured":"Neural machine translation by jointly learning to align and translate\n          \n          \n            \n              Bahdanau\n              D.\n            \n            \n              Cho\n              K.\n            \n            \n              Bengio\n              Y.\n            \n          \n          arXiv\n          \n          2014"},{"key":"B5","doi-asserted-by":"publisher","first-page":"773","DOI":"10.3390\/diagnostics15060773","article-title":"A comparative study of machine learning and deep learning models for automatic Parkinson's disease detection from electroencephalogram signals","volume":"15","author":"Bera","year":"2025","journal-title":"Diagnostics"},{"key":"B6","doi-asserted-by":"crossref","DOI":"10.1016\/j.bnd.2026.07.003","article-title":"JU-PDDv1: a harmonized dataset for Parkinson's disease diagnosis from heterogeneous EEG recordings","author":"Bera","year":"2026","journal-title":"Brain Netw. Disord"},{"key":"B7","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1038\/s41531-025-00913-4","article-title":"Speech and language biomarkers for Parkinson's disease prediction, early diagnosis and progression","volume":"11","author":"Cao","year":"2025","journal-title":"NPJ Parkinsons Dis"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.18112\/openneuro.ds003490.v1.1.0","article-title":"EEG: 3-stim auditory oddball and rest in Parkinson's","author":"Cavanagh","year":"2021","journal-title":"OpenNeuro"},{"key":"B9","doi-asserted-by":"publisher","first-page":"110608","DOI":"10.1016\/j.compbiomed.2025.110608","article-title":"The role of data partitioning on the performance of EEG-based deep learning models in supervised cross-subject analysis: a preliminary study","volume":"196","author":"Del Pup","year":"2025","journal-title":"Comput. Biol. Med"},{"key":"B10","doi-asserted-by":"publisher","first-page":"113389","DOI":"10.1016\/j.eswa.2020.113389","article-title":"Selective opposition based grey wolf optimization","volume":"151","author":"Dhargupta","year":"2020","journal-title":"Expert Syst. Appl"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1499","DOI":"10.3390\/s24051499","article-title":"Machine learning-assisted speech analysis for early detection of Parkinson's disease: a study on speaker diarization and classification techniques","volume":"24","author":"Di Cesare","year":"2024","journal-title":"Sensors"},{"key":"B12","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1016\/S0004-3702(96)00034-3","article-title":"Solving the multiple instance problem with axis-parallel rectangles","volume":"89","author":"Dietterich","year":"1997","journal-title":"Artif. Intell"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/MeMeA.2016.7533761","article-title":"\u201cVoxtester, software for digital evaluation of speech changes in Parkinson disease,\u201d","author":"Dimauro","year":"2016","journal-title":"2016 IEEE international symposium on medical measurements and applications (MeMeA)"},{"key":"B14","volume-title":"Italian Parkinson's Voice and Speech","author":"Dimauro","year":"2019"},{"key":"B15","doi-asserted-by":"publisher","first-page":"796","DOI":"10.1016\/S1474-4422(06)70549-X","article-title":"The mystery of motor asymmetry in Parkinson's disease","volume":"5","author":"Djaldetti","year":"2006","journal-title":"Lancet Neurol"},{"key":"B16","doi-asserted-by":"publisher","first-page":"4323","DOI":"10.3390\/electronics12204323","article-title":"A comprehensive review on multiple instance learning","volume":"12","author":"Fatima","year":"2023","journal-title":"Electronics"},{"key":"B17","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1177\/003754970107600201","article-title":"A new heuristic optimization algorithm: harmony search","volume":"76","author":"Geem","year":"2001","journal-title":"Simulation"},{"key":"B18","doi-asserted-by":"publisher","first-page":"713","DOI":"10.1007\/s12065-019-00279-6","article-title":"A histogram based fuzzy ensemble technique for feature selection","volume":"12","author":"Ghosh","year":"2019","journal-title":"Evol. Intell"},{"key":"B19","doi-asserted-by":"publisher","first-page":"495","DOI":"10.1007\/978-981-16-2543-5_42","article-title":"\u201cPy_FS: a Python package for feature selection using meta-heuristic optimization algorithms,\u201d","author":"Guha","year":"2021","journal-title":"Computational intelligence in pattern recognition"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2651","DOI":"10.1007\/s10462-018-9634-2","article-title":"Recent studies on optimisation method of grey wolf optimiser (GWO): a review (2014-2017)","volume":"52","author":"Hatta","year":"2019","journal-title":"Artif. Intell. Rev"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1316","DOI":"10.3390\/bioengineering10111316","article-title":"Towards a corpus (and language)-independent screening of Parkinson's disease from voice and speech through domain adaptation","volume":"10","author":"Ibarra","year":"2023","journal-title":"Bioengineering"},{"key":"B22","unstructured":"Jaeger\n              H.\n            \n            \n              Trivedi\n              D.\n            \n            \n              Stadtschnitzer\n              M.\n            \n          \n          Mobile Device Voice Recordings at King's College London (mdvr-kcl) from Both Early and Advanced Parkinson's Disease Patients and Healthy Controls\n          \n          2019"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2026.3670586","article-title":"Milfnet: multiple instance learning-based fastslow network with multitask autoencoder for cross-subject EEG emotion recognition","volume":"75","author":"Jin","year":"2026","journal-title":"IEEE Trans. Instrum. Meas"},{"key":"B24","unstructured":"Beyond the signal: medication state effect on EEG-based AI models for Parkinson's disease\n          \n          \n            \n              Kurbatskaya\n              A.\n            \n            \n              L\u00e5der\n              F. N.\n            \n            \n              Nese\n              A. S.\n            \n            \n              Br\u00f8nnick\n              K.\n            \n            \n              Fernandez-Quilez\n              A.\n            \n          \n          arXiv\n          \n          2025"},{"key":"B25","doi-asserted-by":"publisher","first-page":"9297","DOI":"10.1007\/s00521-024-09569-2","article-title":"A novel approach for Parkinson's disease detection using Vold-Kalman order filtering and machine learning algorithms","volume":"36","author":"Latifo\u011flu","year":"2024","journal-title":"Neural Comput. Appl"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1498756","DOI":"10.3389\/fnagi.2024.1498756","article-title":"Global, regional, national epidemiology and trends of Parkinson's disease from 1990 to 2021: findings from the global burden of disease study 2021","volume":"16","author":"Luo","year":"2025","journal-title":"Front. Aging Neurosci"},{"key":"B27","volume-title":"Quantifying Motor Vigour in Parkinson's Disease Using EEG: A Dual Workflow of Orthogonal Multiple Instance Learning and ICA-Based Analysis","author":"Mahdizadeh","year":"2024"},{"key":"B28","doi-asserted-by":"publisher","first-page":"109811","DOI":"10.1016\/j.bspc.2026.109811","article-title":"Parkinson's disease detection from resting-state eeg using a transformer model with multi-head attention explanations","volume":"119","author":"Mancazzo","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"B29","doi-asserted-by":"publisher","first-page":"727","DOI":"10.1016\/j.neubiorev.2016.07.010","article-title":"Motor symptoms in Parkinson's disease: a unified framework","volume":"68","author":"Moustafa","year":"2016","journal-title":"Neurosci. Biobeh. Rev"},{"key":"B30","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/978-3-031-78761-4_1","article-title":"\u201cParkinson's disease detection from resting state EEG using multi-head graph structure learning with gradient weighted graph attention explanations,\u201d","author":"Neves","year":"2024","journal-title":"Machine learning in clinical neuroimaging: 7th international workshop, MLCN 2024, held in conjunction with MICCAI 2024"},{"key":"B31","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/978-3-032-09153-6_1","article-title":"\u201cDeep learning for depression detection using EEG signals: a RNN based approach,\u201d","author":"Paul","year":"2025","journal-title":"Human-centered data analytics: technology for sustainable development. AGC 2025. communications in computer and information science"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.64898\/2026.03.07.710210","article-title":"Characterizing EEG spectro-temporal variability signatures in Alzheimer's and Parkinson's disease","author":"Prieur-Coloma","year":"2026","journal-title":"openRxiv"},{"key":"B33","doi-asserted-by":"publisher","first-page":"556","DOI":"10.1016\/j.bbe.2022.04.002","article-title":"End-to-end deep learning approach for Parkinson's disease detection from speech signals","volume":"42","author":"Quan","year":"2022","journal-title":"Biocybernet. Biomed. Eng"},{"key":"B34","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1007\/978-3-032-13497-4_5","article-title":"\u201cSZ-Net: a hybrid CNN-BiLSTM-BiGRU framework for schizophrenia detection and classification using electroencephalogram signals,\u201d","author":"Saha","year":"2026","journal-title":"Recent Advances in Artificial Neural Networks. Intelligent Systems Reference Library"},{"key":"B35","doi-asserted-by":"publisher","first-page":"34835","DOI":"10.1038\/s41598-025-21407-6","article-title":"Multi-modal deep learning framework for early detection of Parkinson's disease using neurological and physiological data for high-fidelity diagnosis","volume":"15","author":"Sar","year":"2025","journal-title":"Sci. Rep"},{"key":"B36","doi-asserted-by":"publisher","first-page":"10909","DOI":"10.1109\/TNNLS.2023.3245329","article-title":"Probabilistic attention based on gaussian processes for deep multiple instance learning","volume":"35","author":"Schmidt","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.18112\/openneuro.ds004584.v1.0.0","article-title":"Rest eyes open","author":"Singh","year":"2023","journal-title":"OpenNeuro"},{"key":"B38","unstructured":"Towards a generalizable speech marker for Parkinson's disease diagnosis\n          \n          \n            \n              Siniukov\n              M.\n            \n            \n              Xing\n              E.\n            \n            \n              Isfahani\n              S. A.\n            \n            \n              Soleymani\n              M.\n            \n          \n          arXiv\n          \n          2025"},{"key":"B39","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1038\/s41746-024-01217-2","article-title":"Digital biomarkers for precision diagnosis and monitoring in Parkinson's disease","volume":"7","author":"Sun","year":"2024","journal-title":"NPJ Digit. Med"},{"key":"B40","doi-asserted-by":"publisher","first-page":"7189","DOI":"10.3390\/s25237189","article-title":"EEG sensor-based Parkinson's disease detection using a multi-domain feature fusion network","volume":"25","author":"Wang","year":"2025","journal-title":"Sensors"},{"key":"B41","doi-asserted-by":"publisher","first-page":"3231","DOI":"10.1109\/TAFFC.2025.3581388","article-title":"Emotionmil: an end-to-end multiple instance learning framework for emotion recognition from EEG signals","volume":"16","author":"Xiao","year":"2025","journal-title":"IEEE Trans. Affect. Comput"},{"key":"B42","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1007\/s10548-025-01102-5","article-title":"Efficient neural network classification of Parkinson's disease and schizophrenia using resting-state EEG data","volume":"38","author":"Xiong","year":"2025","journal-title":"Brain Topogr"},{"key":"B43","doi-asserted-by":"publisher","first-page":"1586273","DOI":"10.3389\/fnagi.2025.1586273","article-title":"Non-invasive detection of Parkinson's disease based on speech analysis and interpretable machine learning","volume":"17","author":"Xu","year":"2025","journal-title":"Front. Aging Neurosci"}],"container-title":["Frontiers in Aging Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnagi.2026.1924167\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,11,5]],"date-time":"2026-11-05T13:34:38Z","timestamp":1793885678000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnagi.2026.1924167\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,9]]},"references-count":43,"alternative-id":["10.3389\/fnagi.2026.1924167"],"URL":"https:\/\/doi.org\/10.3389\/fnagi.2026.1924167","relation":{},"ISSN":["1663-4365"],"issn-type":[{"value":"1663-4365","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,9]]},"article-number":"1924167"}}