{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T06:46:11Z","timestamp":1776062771917,"version":"3.50.1"},"reference-count":45,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T00:00:00Z","timestamp":1776038400000},"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. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Parkinson\u2019s disease (PD) is a progressive neurodegenerative disorder caused by the loss of dopaminergic neurons in the substantia nigra, presenting with motor and non-motor symptoms in roughly 2\u20133% of the global population above age 60. Early detection is difficult because symptoms are subtle and often indistinguishable from normal aging. The MDS-UPDRS rating scale, the current clinical standard, is time-intensive, subjective, and requires experienced clinicians. Most computational approaches are unimodal and do not use image and structured clinical data in combination.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We propose a hybrid quantum-classical dual-track multimodal network that classifies PD patients and healthy controls from hand-drawn spiral and meander patterns. The first track, the Topological Visual\u2013Spatial Feature Encoder Network (TVSFE), uses a ghost module-based CNN with Cross-Dimensional Attention Bottleneck (CDAB) blocks incorporating coordinate attention, squeeze-and-excitation, and triplet attention, followed by a quantum variational circuit with amplitude embedding. The second track, the Variational Quantum Feature Mapping Network (VQFMN), encodes structured clinical and demographic data through RY rotation gates and strongly entangling layers. Outputs from both tracks are concatenated and passed through fully connected layers for classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>On the HandPD test set, the model achieved 97.28% accuracy, 96.60% precision, 96.62% recall, and 96.54% F1-score, outperforming all CNN, transformer, and ML-based baselines compared. Five-fold cross-validation produced a mean accuracy of 96.58%. On the NewHandPD dataset, accuracy, precision, recall, and F1-score were all 95.45%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The quantum-classical fusion outperforms both single-modality and fully classical variants. Grad-CAM localizes the spatial image regions driving classification and the perturbation-based sensitivity analysis identifies Root Mean Square (RMS) and age as the most influential structured features. Both together make the model\u2019s reasoning traceable at the modality level, which is important for decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1807209","type":"journal-article","created":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T05:47:48Z","timestamp":1776059268000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A quantum\u2013classical dual-track deep learning network for explainable Parkinson\u2019s disease classification"],"prefix":"10.3389","volume":"9","author":[{"given":"S.","family":"Alden Jenish","sequence":"first","affiliation":[{"name":"School of Electronics Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Arushi","family":"Pethkar","sequence":"additional","affiliation":[{"name":"School of Electronics Engineering, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"R.","family":"Karthik","sequence":"additional","affiliation":[{"name":"Centre of Cyber-Physical Systems, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"K.","family":"Suganthi","sequence":"additional","affiliation":[{"name":"Centre of Cyber-Physical Systems, Vellore Institute of Technology","place":["Chennai, India"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,4,13]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"3057","DOI":"10.1088\/0031-9155\/51\/12\/004","article-title":"Artificial neural network classifier for the diagnosis of Parkinson's disease using [99mTc]TRODAT-1 and SPECT","volume":"51","author":"Acton","year":"2006","journal-title":"Phys. Med. Biol."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"104312","DOI":"10.1016\/j.compbiomed.2021.104312","article-title":"Improved motor outcome prediction in Parkinson's disease applying deep learning to DaTscan SPECT images","volume":"132","author":"Adams","year":"2021","journal-title":"Comput. Biol. Med."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"282","DOI":"10.1016\/j.future.2018.11.054","article-title":"A recurrence plot-based approach for Parkinson\u2019s disease identification","volume":"94","author":"Afonso","year":"2019","journal-title":"Futur. Gener. Comput. Syst."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"56837","DOI":"10.1007\/s11042-023-17762-z","article-title":"Parkinson's disease diagnosis from T1 and T2 weighted magnetic resonance images using FBLstmNet architecture","volume":"83","author":"Akram","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"267","DOI":"10.11591\/ijeecs.v30.i1.pp267-275","article-title":"Convolutional neural network for the detection of Parkinson disease based on hand-draw spiral images","volume":"30","author":"Alniemi","year":"2023","journal-title":"Indones. J. Electr. Eng. Comput. Sci."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"2658","DOI":"10.5281\/zenodo.8021545","article-title":"Parkinson\u2019s disease detection using spiral drawings","volume":"8","author":"Anisha","year":"2023","journal-title":"Int. J. Innov. Sci. Res. Technol."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"9276579","DOI":"10.1155\/2022\/9276579","article-title":"Metaheuristics with deep learning-enabled Parkinson's disease diagnosis and classification model","volume":"2022","author":"Bahaddad","year":"2022","journal-title":"J. Healthc. Eng."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"549","DOI":"10.34768\/amcs-2021-0037","article-title":"A hybrid two\u2013stage SqueezeNet and support vector machine system for Parkinson\u2019s disease detection based on handwritten spiral patterns","volume":"31","author":"Bernardo","year":"2021","journal-title":"Int. J. Appl. Math. Comput. Sci."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"2724","DOI":"10.1038\/s41598-021-81209-4","article-title":"Texture-based markers from structural imaging correlate with motor handicap in Parkinson\u2019s disease","volume":"11","author":"Betrouni","year":"2021","journal-title":"Sci. Rep."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"204","DOI":"10.1186\/s12880-023-01169-1","article-title":"Multi-parametric radiomics of conventional T1 weighted and susceptibility-weighted imaging for differential diagnosis of idiopathic Parkinson\u2019s disease and multiple system atrophy","volume":"23","author":"Bu","year":"2023","journal-title":"BMC Med. Imaging"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1038\/s41531-024-00647-9","article-title":"Exploiting macro- and micro-structural brain changes for improved Parkinson\u2019s disease classification from MRI data","volume":"10","author":"Camacho","year":"2024","journal-title":"npj Parkinsons Dis."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"2050044","DOI":"10.1142\/S0129065720500446","article-title":"Morphological characterization of functional brain imaging by isosurface analysis in Parkinson\u2019s disease","volume":"30","author":"Castillo-Barnes","year":"2020","journal-title":"Int. J. Neural Syst."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"866666","DOI":"10.3389\/fnins.2022.866666","article-title":"Combining neuroimaging and omics datasets for disease classification using graph neural networks","volume":"16","author":"Chan","year":"2022","journal-title":"Front. Neurosci."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1297","DOI":"10.3390\/brainsci11101297","article-title":"Screening of Parkinson\u2019s disease using geometric features extracted from spiral drawings","volume":"11","author":"Chandra","year":"2021","journal-title":"Brain Sci."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"310","DOI":"10.1186\/s12967-023-04158-8","article-title":"Detection of mild cognitive impairment in Parkinson\u2019s disease using gradient boosting decision tree models based on multilevel DTI indices","volume":"21","author":"Chen","year":"2023","journal-title":"J. Transl. Med."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"109884","DOI":"10.1016\/j.jneumeth.2023.109884","article-title":"An adaptive weighted attention-enhanced deep convolutional neural network for classification of MRI images of Parkinson's disease","volume":"394","author":"Cui","year":"2023","journal-title":"J. Neurosci. Methods"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"4218","DOI":"10.3390\/math10224218","article-title":"Early detection of Parkinson\u2019s disease using fusion of discrete wavelet transformation and histograms of oriented gradients","volume":"10","author":"Das","year":"2022","journal-title":"Mathematics"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"20089","DOI":"10.1007\/s00521-024-10299-8","article-title":"An innovative approach for parkinson\u2019s disease diagnosis using CNN, NCA, and SVM","volume":"36","author":"Dogan","year":"2024","journal-title":"Neural Comput. & Applic."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"102359","DOI":"10.1016\/j.mex.2023.102359","article-title":"Radiomics for Parkinson's disease classification using advanced texture-based biomarkers","volume":"11","author":"Gore","year":"2023","journal-title":"MethodsX"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"2566","DOI":"10.3390\/math10152566","article-title":"Convolutional neural network-based Parkinson disease classification using SPECT imaging data","volume":"10","author":"Hathaliya","year":"2022","journal-title":"Mathematics"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1691","DOI":"10.3390\/diagnostics13101691","article-title":"Prediction of cognitive decline in Parkinson\u2019s disease using clinical and DAT SPECT imaging features, and hybrid machine learning systems","volume":"13","author":"Hosseinzadeh","year":"2023","journal-title":"Diagnostics"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"4792","DOI":"10.3390\/molecules25204792","article-title":"Classification of the multiple stages of Parkinson\u2019s disease by a deep convolution neural network based on 99mTc-TRODAT-1 SPECT images","volume":"25","author":"Hsu","year":"2020","journal-title":"Molecules"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"2827","DOI":"10.3390\/diagnostics13172827","article-title":"Classification of Parkinson\u2019s disease in patch-based MRI of substantia Nigra","volume":"13","author":"Hussain","year":"2023","journal-title":"Diagnostics"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"795","DOI":"10.1136\/jnnp-2019-322338","article-title":"Parkinson's disease: etiopathogenesis and treatment","volume":"91","author":"Jankovic","year":"2020","journal-title":"J. Neurol. Neurosurg. Psychiatry"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"6382619","DOI":"10.1155\/2021\/6382619","article-title":"Self-attention-guided recurrent neural network and motion perception for intelligent prediction of chronic diseases","volume":"2021","author":"Ma","year":"2021","journal-title":"J. Healthc. Eng."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1186\/s12880-024-01335-z","article-title":"An improved method for diagnosis of Parkinson\u2019s disease using deep learning models enhanced with metaheuristic algorithm","volume":"24","author":"Majhi","year":"2024","journal-title":"BMC Med. Imaging"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"8383","DOI":"10.1007\/s00521-024-09516-1","article-title":"Quantum mayfly optimization based feature subset selection with hybrid CNN for biomedical Parkinson\u2019s disease diagnosis","volume":"36","author":"Mansour","year":"2024","journal-title":"Neural Comput. & Applic."},{"key":"ref28","volume-title":"Parkinson\u2019s Disease Detection Using Spiral Images (Hand Drawings)","author":"Medasani","year":"2023"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"107198","DOI":"10.1016\/j.bspc.2024.107198","article-title":"Elasticnet-based vision transformers for early detection of Parkinson\u2019s disease","volume":"101","author":"\u00d6zdemir","year":"2025","journal-title":"Biomed. Signal Process. Control."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"20708","DOI":"10.1038\/s41598-024-71860-y","article-title":"Multi-modality radiomics of conventional T1 weighted and diffusion tensor imaging for differentiating Parkinson\u2019s disease motor subtypes in early-stages","volume":"14","author":"Panahi","year":"2024","journal-title":"Sci. Rep."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.37394\/232018.2021.9.1","article-title":"\u0397yper-sinh-convolutional neural network for early detection of Parkinson\u2019s disease from spiral drawings","volume":"9","author":"Parisi","year":"2021","journal-title":"WSEAS Trans. Comput. Res."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"259","DOI":"10.3390\/bios14050259","article-title":"Parkinson\u2019s disease recognition using decorrelated convolutional neural networks: addressing imbalance and scanner bias in rs-fMRI data","volume":"14","author":"Patil","year":"2024","journal-title":"Biosensors"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1016\/j.cmpb.2016.08.005","article-title":"A new computer vision-based approach to aid the diagnosis of Parkinson's disease","volume":"136","author":"Pereira","year":"","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref34","first-page":"340","author":"Pereira","year":""},{"key":"ref35","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1186\/s42466-020-00092-y","article-title":"A machine learning-based classification approach on Parkinson\u2019s disease diffusion tensor imaging datasets","volume":"2","author":"Prasuhn","year":"2020","journal-title":"Neurol. Res. Pract."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"11","DOI":"10.52783\/jns.v14.1725","article-title":"Parkinson\u2019s disease prediction with spiral drawings and wave frequency using deep conformal neural networks","volume":"14","author":"SasiRekha","year":"2025","journal-title":"J. Neonatal Surg."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"107274","DOI":"10.1016\/j.bspc.2024.107274","article-title":"Multimodal Parkinson\u2019s disease classification using voice signals and hand-drawn images based on deep learning enabled by chronological tangent search optimization","volume":"102","author":"Shanthini","year":"2025","journal-title":"Biomed. Signal Process. Control."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"7921","DOI":"10.1007\/s00330-024-10886-2","article-title":"Radiomics score derived from T1-w\/T2-w ratio image can predict motor symptom progression in Parkinson\u2019s disease","volume":"34","author":"Shimozono","year":"2024","journal-title":"Eur. Radiol."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"13813","DOI":"10.1038\/s41598-023-40199-4","article-title":"Detection of Parkinson disease using multiclass machine learning approach","volume":"14","author":"Srinivasan","year":"2024","journal-title":"Sci. Rep."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"e50117","DOI":"10.2196\/50117","article-title":"Machine learning models for Parkinson disease: systematic review","volume":"12","author":"Tabashum","year":"2024","journal-title":"JMIR Med. Inform."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"105140","DOI":"10.1016\/j.bspc.2023.105140","article-title":"PDDS: a software for the early diagnosis of Parkinson\u2019s disease from MRI and DaT scan images using detection and segmentation algorithms","volume":"86","author":"Tassew","year":"2023","journal-title":"Biomed. Signal Process. Control"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"108066","DOI":"10.1016\/j.cmpb.2024.108066","article-title":"LSTM-CNN: an efficient diagnostic network for Parkinson's disease utilizing dynamic handwriting analysis","volume":"247","author":"Wang","year":"2024","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"2686","DOI":"10.1049\/ipr2.12820","article-title":"A coordinate attention enhanced swin transformer for handwriting recognition of Parkinson's disease","volume":"17","author":"Wang","year":"2023","journal-title":"IET Image Process."},{"key":"ref44","doi-asserted-by":"publisher","first-page":"109019","DOI":"10.1016\/j.jneumeth.2020.109019","article-title":"Classification of Parkinson's disease based on multi-modal features and stacking ensemble learning","volume":"350","author":"Yang","year":"2021","journal-title":"J. Neurosci. Methods"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"104085","DOI":"10.1016\/j.jbi.2022.104085","article-title":"Early diagnosis of Parkinson's disease using continuous convolution network: handwriting recognition based on off-line hand drawing without template","volume":"130","author":"Zhu","year":"2022","journal-title":"J. Biomed. Inform."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1807209\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T05:47:49Z","timestamp":1776059269000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1807209\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,13]]},"references-count":45,"alternative-id":["10.3389\/frai.2026.1807209"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1807209","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,13]]},"article-number":"1807209"}}