{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T22:31:58Z","timestamp":1793485918156,"version":"4.5.0"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T00:00:00Z","timestamp":1772150400000},"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. Digit. Health"],"abstract":"<jats:p>Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by motor and non-motor impairments, where early diagnosis remains challenging due to reliance on subjective clinical assessments. Recent artificial intelligence (AI)-based approaches have demonstrated promise in identifying subtle PD biomarkers from individual modalities such as speech, gait, and handwriting; however, unimodal systems often fail to capture the heterogeneity of the disease and provide limited interpretability. To address these limitations, this study proposes a multimodal deep learning framework that integrates handwriting, gait, and speech modalities using an early feature fusion strategy for robust and interpretable PD detection. Each modality is processed through a dedicated feature extraction pipeline using deep neural networks, followed by static feature concatenation and classification using an XGBoost model. Model transparency is enhanced using explainable AI (XAI) techniques, including SHapley Additive exPlanations (SHAP) and Gradient-weighted Class Activation Mapping (Grad-CAM), enabling clinical interpretability of modality- and feature-level contributions. Experimental evaluation on benchmark datasets demonstrates that the proposed trimodal fusion model achieves an accuracy of 92%, outperforming unimodal handwriting (91%), gait (90%), and speech (74%) models. The fusion framework attains a macro F1-score of 0.89, an area under the ROC curve (AUC) of 0.95, and an average precision (AP) of 0.96, indicating strong discriminative capability and robustness. Confusion matrix analysis reveals balanced sensitivity (90%) and specificity (89%) across classes. Explainability analysis confirms that handwriting tremor patterns, gait force asymmetries, and speech spectral instabilities are key contributors to PD prediction. These results highlight the effectiveness of explainable multimodal AI in delivering accurate, reliable, and clinically interpretable solutions for early PD detection.<\/jats:p>","DOI":"10.3389\/fdgth.2026.1771281","type":"journal-article","created":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T06:51:33Z","timestamp":1772175093000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":10,"title":["Explainable multimodal feature fusion networks for Parkinson's disease prediction"],"prefix":"10.3389","volume":"8","author":[{"given":"Abishek","family":"Ravichandran","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tamilarasi","family":"Kathirvel Murugan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Logeswari","family":"Govindaraj","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vishal","family":"M","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,27]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"79549","DOI":"10.1109\/ACCESS.2024.3408680","article-title":"Enhancing Parkinson\u2019s disease diagnosis through stacking ensemble-based machine learning approach","volume":"12","author":"Al-Tam","year":"2024","journal-title":"IEEE Access"},{"key":"B2","doi-asserted-by":"publisher","first-page":"27821","DOI":"10.1109\/ACCESS.2020.2968177","article-title":"Deep feature extraction from the vocal vectors using sparse autoencoders for Parkinson\u2019s classification","volume":"8","author":"Xiong","year":"2020","journal-title":"IEEE Access"},{"key":"B3","doi-asserted-by":"publisher","first-page":"147635","DOI":"10.1109\/ACCESS.2020.3016062","article-title":"Early detection of Parkinson\u2019s disease using deep learning and machine learning","volume":"8","author":"Wang","year":"2020","journal-title":"IEEE Access"},{"key":"B4","doi-asserted-by":"publisher","first-page":"127004","DOI":"10.1109\/ACCESS.2020.3007140","article-title":"Parkinson data analysis and prediction system using multi-variant stacked auto encoder","volume":"8","author":"Nagasubramanian","year":"2020","journal-title":"IEEE Access"},{"key":"B5","doi-asserted-by":"publisher","first-page":"135600","DOI":"10.1109\/ACCESS.2023.3337804","article-title":"Multi-model fusion of CNNs for identification of Parkinson\u2019s disease using (7) handwritten samples","volume":"11","author":"Naz","year":"2023","journal-title":"IEEE Access"},{"key":"B6","doi-asserted-by":"publisher","first-page":"35482","DOI":"10.1109\/ACCESS.2020.2974008","article-title":"A spectrogram-based deep feature assisted computer-aided diagnostic system for Parkinson\u2019s disease","volume":"8","author":"Zahid","year":"2020","journal-title":"IEEE Access"},{"key":"B7","doi-asserted-by":"publisher","first-page":"115540","DOI":"10.1109\/ACCESS.2019.2936564","article-title":"Deep learning-based Parkinson\u2019s disease classification using vocal feature sets","volume":"7","author":"Gunduz","year":"2019","journal-title":"IEEE Access"},{"key":"B8","doi-asserted-by":"publisher","first-page":"3937","DOI":"10.1109\/TNSRE.2023.3314100","article-title":"An interpretable deep learning optimized wearable daily detection system for Parkinson\u2019s disease","volume":"31","author":"Chen","year":"2023","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"key":"B9","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1007\/s11063-025-11735-z","article-title":"Parkinson\u2019s detection from gait time series classification using modified metaheuristic optimized long short-term memory","volume":"57","author":"Markovic","year":"2025","journal-title":"Neural Process Lett"},{"key":"B10","doi-asserted-by":"publisher","first-page":"73","DOI":"10.3390\/diagnostics15010073","article-title":"Synchronous analysis of speech production and lips movement to detect Parkinson\u2019s disease using deep learning methods","volume":"15","author":"R\u00edos-Urrego","year":"2025","journal-title":"Diagnostics"},{"key":"B11","doi-asserted-by":"publisher","first-page":"19033","DOI":"10.1109\/ACCESS.2022.3150774","article-title":"Early detection of Parkinson\u2019s disease by neural network models","volume":"10","author":"Lin","year":"2022","journal-title":"IEEE Access"},{"key":"B12","doi-asserted-by":"publisher","first-page":"56936","DOI":"10.1109\/ACCESS.2024.3387342","article-title":"Parkinson\u2019s disease detection from online handwriting based on beta-elliptical approach and fuzzy perceptual detector","volume":"12","author":"Allebawi","year":"2024","journal-title":"IEEE Access"},{"key":"B13","doi-asserted-by":"publisher","first-page":"316","DOI":"10.1007\/s44230-024-00069-z","article-title":"Single and multi-modal analysis for Parkinson\u2019s disease to detect its underlying factors","volume":"4","author":"Islam","year":"2024","journal-title":"Hum Centric Intell Syst"},{"key":"B14","doi-asserted-by":"publisher","first-page":"4","DOI":"10.3390\/diagnostics15010004","article-title":"Parkinson\u2019s disease prediction: an attention-based multimodal fusion framework using handwriting and clinical data","volume":"15","author":"Benredjem","year":"2025","journal-title":"Diagnostics"},{"key":"B15","doi-asserted-by":"publisher","first-page":"4459","DOI":"10.1109\/TNSRE.2023.3330643","article-title":"Multi-level graph neural network with sparsity pooling for recognizing Parkinson\u2019s disease","volume":"31","author":"Zhang","year":"2023","journal-title":"EEE Trans Neural Syst Rehabil Eng"},{"key":"B16","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","year":"2025","journal-title":"Information"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2399","DOI":"10.1109\/TNSRE.2023.3277749","article-title":"Multi-modal deep learning diagnosis of Parkinson\u2019s disease\u2014a systematic review","volume":"31","author":"Skaramagkas","year":"2023","journal-title":"EEE Trans Neural Syst Rehabil Eng"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2293","DOI":"10.1109\/TNSRE.2024.3416446","article-title":"Characterizing disease progression in Parkinson\u2019s disease from videos of the finger tapping test","volume":"32","author":"Guar\u00edn","year":"2024","journal-title":"EEE Trans Neural Syst Rehabil Eng"},{"key":"B19","first-page":"29","article-title":"Assessment of Parkinson\u2019s disease severity using gait data: a deep learning-based multimodal approach","volume-title":"Digital Health and Wireless Solutions. NCDHWS 2024. Communications in Computer and Information Science (Vol. 2084)","author":"Faiem","year":"2024"},{"key":"B20","doi-asserted-by":"publisher","first-page":"20941","DOI":"10.1038\/s41598-024-70165-4","article-title":"Enhancing early Parkinson\u2019s disease detection through multimodal deep learning and explainable AI: insights from the PPMI database","volume":"14","author":"Dentamaro","year":"2024","journal-title":"Sci Rep"},{"key":"B21","volume-title":"Explainable Prediction of Parkinson\u2019s Disease in a Large Multimodal Database","author":"Camacho Camacho","year":"2023"},{"key":"B22","doi-asserted-by":"publisher","first-page":"103541","DOI":"10.1016\/j.retram.2025.103541","article-title":"Explainable AI for Parkinson\u2019s disease prediction: a machine learning approach with interpretable models","volume":"73","author":"Esan","year":"2025","journal-title":"Curr Res Transl Med"},{"key":"B23","doi-asserted-by":"publisher","first-page":"110363","DOI":"10.1016\/j.jneumeth.2025.110363","article-title":"PIDGN: an explainable multimodal deep learning framework for early prediction of Parkinson\u2019s disease","volume":"415","author":"Li","year":"2025","journal-title":"J Neurosci Methods"},{"key":"B24","doi-asserted-by":"publisher","first-page":"17457","DOI":"10.1109\/ACCESS.2025.3533703","article-title":"Design of an early prediction model for Parkinson\u2019s disease using machine learning","volume":"13","author":"Velu","year":"2025","journal-title":"IEEE Access"},{"key":"B25","doi-asserted-by":"publisher","first-page":"66","DOI":"10.53070\/bbd.1350356","article-title":"Predicting Parkinson\u2019s disease progression: a non-invasive method leveraging voice inputs","volume":"8","author":"Hassan","year":"2023","journal-title":"Comput Sci"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1771281\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T06:51:34Z","timestamp":1772175094000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1771281\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,27]]},"references-count":25,"alternative-id":["10.3389\/fdgth.2026.1771281"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1771281","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,27]]},"article-number":"1771281"}}