{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T06:11:27Z","timestamp":1760076687877,"version":"build-2065373602"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T00:00:00Z","timestamp":1760054400000},"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. Comput. Neurosci."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Parkinson\u2019s disease (PD) is the fastest-growing neurodegenerative disorder, with subtle gait changes such as reduced vertical ground-reaction forces (VGRF) often preceding motor symptoms. These gait abnormalities, measurable via wearable VGRF sensors, offer a non-invasive means for early PD detection. However, current computational approaches often suffer from redundant features and class imbalance, limiting both accuracy and generalizability.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We propose CRISP (Correlation-filtered Recursive Feature Elimination and Integration of SMOTE Pipeline for Gait-Based Parkinson\u2019s Disease Screening), a lightweight multistage framework that sequentially applies correlation-based feature pruning, recursive feature elimination (RFE), and Synthetic Minority Oversampling Technique (SMOTE) based class balancing. To ensure clinically meaningful evaluation, a novel subject-wise protocol was also introduced that assigns one prediction per individual enhancing patient-level variability capture and better aligning with diagnostic workflows. Using 306 VGRF recordings (93 PD, 76 controls), five classifiers, i.e., \ufeffk-Nearest Neighbours (KNN), Decision Tree (DT), Random Forest (RF), Gradient boosting (GB), and Extreme Gradient Boosting (XGBoost) were evaluated for both binary PD detection and multiclass severity grading.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>CRISP consistently improved performance across all models under 5-fold cross-validation. XGBoost achieved the highest performance, increasing subject-wise PD detection accuracy from 96.1\u202f\u00b1\u202f0.8% to 98.3\u202f\u00b1\u202f0.8%, and severity grading accuracy from 96.2\u202f\u00b1\u202f0.7% to 99.3\u202f\u00b1\u202f0.5%.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>CRISP is the first VGRF-based pipeline to combine correlation-filtered feature pruning, recursive feature elimination, and SMOTE to enhance PD detection performance, while also introducing a subject-wise evaluation protocol that captures patient-level variability for truly personalized diagnostics. These twin novelties deliver clinically significant gains and lay the foundation for real-time, on-device PD detection and severity monitoring.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fncom.2025.1660963","type":"journal-article","created":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T05:28:37Z","timestamp":1760074117000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["CRISP: a correlation-filtered recursive feature elimination and integration of SMOTE pipeline for gait-based Parkinson\u2019s disease screening"],"prefix":"10.3389","volume":"19","author":[{"given":"Namra","family":"Afzal","sequence":"first","affiliation":[]},{"given":"Javaid","family":"Iqbal","sequence":"additional","affiliation":[]},{"given":"Asim","family":"Waris","sequence":"additional","affiliation":[]},{"given":"Muhammad Jawad","family":"Khan","sequence":"additional","affiliation":[]},{"given":"Fawwaz","family":"Hazzazi","sequence":"additional","affiliation":[]},{"given":"Hasnain","family":"Ali","sequence":"additional","affiliation":[]},{"given":"Muhammad Adeel","family":"Ijaz","sequence":"additional","affiliation":[]},{"given":"Syed Omer","family":"Gilani","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,10,10]]},"reference":[{"key":"ref9001","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1186\/s12984-024-01458-y","article-title":"Multibody dynamics-based musculoskeletal modeling for gait analysis: a systematic review","volume":"21","author":"Abdullah","year":"2024","journal-title":"J NeuroEngineering Rehabil"},{"key":"ref1","doi-asserted-by":"publisher","first-page":"1377165","DOI":"10.3389\/fdgth.2024.1377165","article-title":"A comparative study in class imbalance mitigation when working with physiological signals","volume":"6","author":"Abdulsadig","year":"2024","journal-title":"Front. Digit. Health"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"e0175951","DOI":"10.1371\/journal.pone.0175951","article-title":"Vertical ground reaction force marker for Parkinson\u2019s disease","volume":"12","author":"Alam","year":"2017","journal-title":"PLoS One"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/LSENS.2020.2994938","article-title":"Machine learning algorithm for gait analysis and classification on early detection of Parkinson","volume":"4","author":"Alkhatib","year":"2020","journal-title":"IEEE Sens. Lett."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1080\/17582024.2025.2510842","article-title":"Gait analysis for Parkinson\u2019s disease using multiscale entropy","author":"Amisola","year":"2025","journal-title":"Neurodegener. Dis. Manag."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"602","DOI":"10.1016\/j.amjmed.2017.11.051","article-title":"Gait disorders","volume":"131","author":"Baker","year":"2018","journal-title":"Am. J. Med."},{"key":"ref6","author":"Barber\u00e0","year":"2024"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"2284","DOI":"10.1016\/S0140-6736(21)00218-X","article-title":"Parkinson\u2019s disease","volume":"397","author":"Bloem","year":"2021","journal-title":"Lancet"},{"key":"ref8","author":"Brindha","year":"2022"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-024-72648-w","article-title":"Convolutional neural network based detection of early stage Parkinson\u2019s disease using the six minute walk test","volume":"14","author":"Choi","year":"2024","journal-title":"Sci. Rep."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1572","DOI":"10.3390\/s24051572","article-title":"Machine learning and digital biomarkers can detect early stages of neurodegenerative diseases","volume":"24","author":"Chudzik","year":"2024","journal-title":"Sens"},{"key":"ref11","doi-asserted-by":"crossref","first-page":"175","DOI":"10.18653\/v1\/W16-2923","article-title":"An information foraging approach to determining the number of relevant features","volume-title":"Proceedings of the 15th workshop on biomedical natural language processing","author":"Connolly","year":"2016"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"2205","DOI":"10.1109\/TNSRE.2023.3269569","article-title":"Static-dynamic temporal networks for Parkinson\u2019s disease detection and severity prediction","volume":"31","author":"Dong","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"103301","DOI":"10.1016\/j.humov.2024.103301","article-title":"Effects of freezing of gait on vertical ground reaction force in Parkinson\u2019s disease","volume":"98","author":"Etoom","year":"2024","journal-title":"Hum. Mov. Sci."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1002\/mds.20507","article-title":"Treadmill walking as an external pacemaker to improve gait rhythm and stability in Parkinson\u2019s disease","volume":"20","author":"Frenkel-Toledo","year":"2005","journal-title":"Mov. Disord."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"1858","DOI":"10.1080\/10255842.2023.2263125","article-title":"Diagnosis of Parkinson\u2019s disease based on voice signals using SHAP and hard voting ensemble method","volume":"27","author":"Ghaheri","year":"2024","journal-title":"Comput. Methods Biomech. Biomed. Engin."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"e215","DOI":"10.1161\/01.cir.101.23.e215","article-title":"Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals","volume":"101","author":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"ref17","author":"Hausdorff","year":"2008"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"107659","DOI":"10.1016\/j.bspc.2025.107659","article-title":"Gait analysis in mixed reality for Parkinson\u2019s disease assessment","volume":"106","author":"Hemmerling","year":"2025","journal-title":"Biomed. Signal Process. Control."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-019-38748-8","article-title":"Explaining the unique nature of individual gait patterns with deep learning","volume":"9","author":"Horst","year":"2019","journal-title":"Sci. Rep."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1016\/j.cmpb.2017.04.007","article-title":"An automatic non-invasive method for Parkinson\u2019s disease classification","volume":"145","author":"Joshi","year":"2017","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref22","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4614-6849-3","volume-title":"Applied predictive Modeling","author":"Kuhn","year":"2013"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"596","DOI":"10.2174\/1871527320666211006142100","article-title":"Tremor and rigidity in patients with Parkinson\u2019s disease: emphasis on epidemiology, pathophysiology and contributing factors","volume":"21","author":"Kumar","year":"2021","journal-title":"CNS Neurol. Disord. Drug Targets"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"7221","DOI":"10.1007\/s10489-020-02182-5","article-title":"A dual-branch model for diagnosis of Parkinson\u2019s disease based on the independent and joint features of the left and right gait","volume":"51","author":"Liu","year":"2021","journal-title":"Appl. Intell."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"143","DOI":"10.1016\/j.eswa.2019.113075","article-title":"Deep 1D-convnet for accurate Parkinson disease detection and severity prediction from gait","volume":"143","author":"Maachi","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-024-83357-9","article-title":"Gait-based Parkinson\u2019s disease diagnosis and severity classification using force sensors and machine learning","volume":"15","author":"Navita","year":"2025","journal-title":"Sci. Rep."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"200","DOI":"10.3934\/Neuroscience.2023017","article-title":"Depletion of dopamine in Parkinson\u2019s disease and relevant therapeutic options: a review of the literature","volume":"10","author":"Ramesh","year":"2023","journal-title":"AIMS Neurosci."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"2003","DOI":"10.3390\/DIAGNOSTICS12082003","article-title":"Imperative role of machine learning algorithm for detection of Parkinson\u2019s disease: review, challenges and recommendations","volume":"12","author":"Rana","year":"2022","journal-title":"Diagn."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"122","DOI":"10.1016\/j.procs.2024.04.015","article-title":"A comparative analysis of advanced machine learning algorithms to diagnose Parkinson\u2019s disease","volume":"235","author":"Roy","year":"2024","journal-title":"Procedia Comput. Sci."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"21362","DOI":"10.1109\/JSEN.2022.3210773","article-title":"Machine learning for microcontroller-class hardware: a review","volume":"22","author":"Saha","year":"2022","journal-title":"IEEE Sensors J."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"768575","DOI":"10.3389\/fnhum.2022.768575","article-title":"Review\u2014emerging portable technologies for gait analysis in neurological disorders","volume":"16","author":"Salchow-H\u00f6mmen","year":"2022","journal-title":"Front. Hum. Neurosci."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1186\/s12984-021-00958-5","article-title":"Prediction and detection of freezing of gait in Parkinson\u2019s disease from plantar pressure data using long short-term memory neural-networks","volume":"18","author":"Shalin","year":"2021","journal-title":"J. Neuro Engineering Rehabil."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1274575","DOI":"10.3389\/fncom.2023.1274575","article-title":"Monitoring time domain characteristics of Parkinson\u2019s disease using 3D memristive neuromorphic system","volume":"17","author":"Siddique","year":"2023","journal-title":"Front. Comput. Neurosci."},{"key":"ref34","author":"Srinivasarao","year":"2024"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"587057","DOI":"10.3389\/fphys.2020.587057","article-title":"Parkinson\u2019s disease diagnosis and severity assessment using ground reaction forces and neural networks","volume":"11","author":"Veeraragavan","year":"2020","journal-title":"Front. Physiol."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1007\/s11571-022-09925-9","article-title":"Gait classification for early detection and severity rating of Parkinson\u2019s disease based on hybrid signal processing and machine learning methods","volume":"18","author":"Wang","year":"2024","journal-title":"Cogn. Neurodyn."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"1248","DOI":"10.1111\/j.1460-9568.2005.04298.x","article-title":"Dual tasking, gait rhythmicity, and Parkinson\u2019s disease: which aspects of gait are attention demanding?","volume":"22","author":"Yogev","year":"2005","journal-title":"Eur. J. Neurosci."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1016\/j.neunet.2018.12.012","article-title":"Classification of gait patterns between patients with Parkinson\u2019s disease and healthy controls using phase space reconstruction (PSR), empirical mode decomposition (EMD) and neural networks","volume":"111","author":"Zeng","year":"2019","journal-title":"Neural Netw."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"13830","DOI":"10.1038\/s41598-025-98571-2","article-title":"Efficient human activity recognition on edge devices using DeepConv LSTM architectures","volume":"15","author":"Zhou","year":"2025","journal-title":"Sci. Rep."}],"container-title":["Frontiers in Computational Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2025.1660963\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T05:28:39Z","timestamp":1760074119000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2025.1660963\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,10]]},"references-count":39,"alternative-id":["10.3389\/fncom.2025.1660963"],"URL":"https:\/\/doi.org\/10.3389\/fncom.2025.1660963","relation":{},"ISSN":["1662-5188"],"issn-type":[{"value":"1662-5188","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,10]]},"article-number":"1660963"}}