{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T15:06:53Z","timestamp":1772118413163,"version":"3.50.1"},"reference-count":35,"publisher":"Termedia Sp. z.o.o.","issue":"1","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Pol J Radiol"],"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>It is still hard to diagnose Parkinson\u2019s disease (PD) early and correctly, since the motor symptoms are often relatively mild and there are no unique diagnostic tools. The goal of this project was to create and test a hybrid deep learning model that can accurately classify PD by using hand-drawn spiral and waveform pictures.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Material and methods<\/jats:title>\n                    <jats:p>A novel two-channel hybrid model was created, where the input representation combines normalised greyscale features and Canny edge features to capture both spatial and structural stroke patterns from patient drawings. The model combines a convolutional neural network (CNN) with hand-crafted grey-level co-occurrence matrix (GLCM) features to enhance its performance and make it easier to understand. We trained and evaluated three different models: a baseline CNN, a fusion CNN + GLCM, and a fine-tuned ResNet-50. We did this on both the original and pre-processed datasets.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The hybrid CNN + GLCM model that used pre-processing had the best classification accuracy at 97.02% and worked well on datasets that were not used to train it. Statistical studies validated the importance of enhancements in performance relative to baseline models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The suggested technique provides a straightforward, comprehensible, and efficient approach for PD screening using easily administered drawing exercises. Its great precision and low equipment needs make it a good candidate for use in real-world clinical settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.5114\/pjr\/213689","type":"journal-article","created":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T13:57:33Z","timestamp":1772114253000},"page":"75-94","source":"Crossref","is-referenced-by-count":0,"title":["Interpretable Hybrid Deep Learning Model for Parkinson\u2019s Disease Screening from Hand-Sketch Spiral and Waveform Images"],"prefix":"10.5114","volume":"91","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6715-8596","authenticated-orcid":true,"given":"Garima","family":"Verma","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"3061","published-online":{"date-parts":[[2026,2,18]]},"reference":[{"key":"ref-1677193","doi-asserted-by":"crossref","unstructured":"Zeng R, Wang J, Zheng C, Jiang R, Tong S, Wu H, et al. Lack of causal associations of inflammatory bowel disease with Parkinson\u2019s disease and other neurodegenerative disorders. Mov Disord 2023; 38: 1082-1088.","DOI":"10.1002\/mds.29386"},{"key":"ref-1677194","doi-asserted-by":"crossref","unstructured":"Essa MM. Handbook of neurodegenerative disorders. Singapore: Springer; 2024.","DOI":"10.1007\/978-981-99-7557-0"},{"key":"ref-1677195","doi-asserted-by":"publisher","DOI":"10.1186\/s13024-023-00622-7"},{"key":"ref-1677196","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2023.1137366"},{"key":"ref-1677197","doi-asserted-by":"crossref","unstructured":"Al-Jabbar M, Alshahrani M, Senan EM, Abunadi I, Almalki SA, Alshari EA. Hybrid techniques of multi-CNN and ensemble learning to analyze handwritten spiral and wave drawing for diagnosing Parkinson\u2019s disease. Comput Model Eng Sci 2025; 143: 2429-2457.","DOI":"10.32604\/cmes.2025.063938"},{"key":"ref-1677198","doi-asserted-by":"crossref","unstructured":"Aouraghe I, Khaissidi G, Mrabti M. A literature review of online handwriting analysis to detect Parkinson\u2019s disease at an early stage. Multimed Tools Appl 2023; 82: 11923-11948.","DOI":"10.1007\/s11042-022-13759-2"},{"key":"ref-1677199","doi-asserted-by":"crossref","unstructured":"Thakur M, Dhanalakshmi S, Kuresan H, Senthil R, Narayanamoorthi R, Lai KW. Automated restricted Boltzmann machine classifier for early diagnosis of Parkinson\u2019s disease using digitized spiral drawings. J Ambient Intell Humaniz Comput 2023; 14: 175-189.","DOI":"10.1007\/s12652-022-04361-3"},{"key":"ref-1677200","doi-asserted-by":"publisher","DOI":"10.1186\/s12984-023-01219-3"},{"key":"ref-1677201","doi-asserted-by":"publisher","DOI":"10.1016\/j.arr.2023.102013"},{"key":"ref-1677202","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13111924"},{"key":"ref-1677203","doi-asserted-by":"crossref","unstructured":"Wang Y, He N, Zhang C, Zhang Y, Wang C, Huang P, et al. An automatic interpretable deep learning pipeline for accurate Parkinson\u2019s disease diagnosis using quantitative susceptibility mapping and T1-weighted images. Hum Brain Mapp 2023; 44: 4426-4438.","DOI":"10.1002\/hbm.26399"},{"key":"ref-1677204","doi-asserted-by":"crossref","unstructured":"Ayaz Z, Naz S, Khan NH, Razzak I, Imran M. Automated methods for diagnosis of Parkinson\u2019s disease and predicting severity level. Neural Comput Appl 2023; 35: 14499-14534.","DOI":"10.1007\/s00521-021-06626-y"},{"key":"ref-1677205","doi-asserted-by":"crossref","unstructured":"Ma YW, Chen JL, Chen YJ, Lai YH. Explainable deep learning architecture for early diagnosis of Parkinson\u2019s disease. Soft Comput 2023; 27: 2729-2738.","DOI":"10.1007\/s00500-021-06170-w"},{"key":"ref-1677206","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.1485"},{"key":"ref-1677207","doi-asserted-by":"crossref","unstructured":"Varalakshmi P, Priya BT, Rithiga BA, Bhuvaneaswari R, Sundar RS. Diagnosis of Parkinson\u2019s disease from hand drawing utilizing hybrid models. Parkinsonism Relat Disord 2022; 105: 24-31.","DOI":"10.1016\/j.parkreldis.2022.10.020"},{"key":"ref-1677208","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-86290-8_3"},{"key":"ref-1677209","doi-asserted-by":"crossref","unstructured":"Alissa M, Lones MA, Cosgrove J, Alty JE, Jamieson S, Smith SL, Vallejo M. Parkinson\u2019s disease diagnosis using convolutional neural networks and figure-copying tasks. Neural Comput Appl 2022; 34: 1433-1453.","DOI":"10.1007\/s00521-021-06469-7"},{"key":"ref-1677210","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.12739"},{"key":"ref-1677211","doi-asserted-by":"crossref","unstructured":"Prasanna ML, Kumar C, Mishra RK. Enhanced Parkinson\u2019s diagnosis through ensemble learning with stacking and cross-validation. Procedia Comput Sci 2025; 259: 250-259.","DOI":"10.1016\/j.procs.2025.03.326"},{"key":"ref-1677212","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024"},{"key":"ref-1677213","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc"},{"key":"ref-1677214","doi-asserted-by":"publisher","DOI":"10.1016\/j.sna.2025.116287"},{"key":"ref-1677215","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.107274"},{"key":"ref-1677216","doi-asserted-by":"publisher","DOI":"10.1016\/j.medengphy.2025.104333"},{"key":"ref-1677217","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2022"},{"key":"ref-1677218","doi-asserted-by":"publisher","DOI":"10.2174\/0118741207294056240322075602"},{"key":"ref-1677219","doi-asserted-by":"publisher","DOI":"10.2174\/0118749445302023240520111802"},{"key":"ref-1677220","unstructured":"Augmented Hand-Drawn Data for Parkinson\u2019s Disease. 2025. Available at: https:\/\/www.kaggle.com\/datasets\/abdulkhalekmugahed\/augmented-hand-drawn-data-for-parkinsons-disease\/versions\/1 (accessed 20 July 2025)."},{"key":"ref-1677221","unstructured":"Parkinson\u2019s Drawings. 2025. Available at: https:\/\/www.kaggle.com\/datasets\/kmader\/parkinsons-drawings\/data (accessed 20 July 2025)."},{"key":"ref-1677222","doi-asserted-by":"publisher","DOI":"10.5114\/pjr\/189412"},{"key":"ref-1677223","doi-asserted-by":"publisher","DOI":"10.1016\/j.iswa.2025.200492"},{"key":"ref-1677224","doi-asserted-by":"crossref","unstructured":"Maithani A, Verma G. Hybrid model with improved score level fusion for heart disease classification. Multimed Tools Appl 2024; 83: 54951-54987.","DOI":"10.1007\/s11042-023-17599-6"},{"key":"ref-1677225","doi-asserted-by":"publisher","DOI":"10.1016\/j.buildenv.2025.112874"},{"key":"ref-1677226","doi-asserted-by":"crossref","unstructured":"Thakkar B. Continuous variable analyses: student\u2019s t-test, Mann-Whitney U test, Wilcoxon signed-rank test. In: Eltorai AEM, Bakal JA, Gibson CM, editors. Handbook for designing and conducting clinical and translational research. Translational Cardiology. Boston, MA: Academic Press; 2025: 165-167.","DOI":"10.1016\/B978-0-323-91790-2.00052-6"},{"key":"ref-1677227","doi-asserted-by":"publisher","DOI":"10.1016\/j.suscom.2025.101115"}],"container-title":["Polish Journal of Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.polradiol.com\/pdf-213689-135948","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.polradiol.com\/pdf-213689-135948","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T13:57:39Z","timestamp":1772114259000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.polradiol.com\/Interpretable-Hybrid-Deep-Learning-Model-for-Parkinson-s-Disease-Screening-from-Hand,213689,0,2.html"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,18]]},"references-count":35,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026]]}},"alternative-id":["213689"],"URL":"https:\/\/doi.org\/10.5114\/pjr\/213689","relation":{},"ISSN":["1899-0967"],"issn-type":[{"value":"1899-0967","type":"print"}],"subject":[],"published":{"date-parts":[[2026,2,18]]}}}