{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T21:38:27Z","timestamp":1790113107913,"version":"4.0.1"},"reference-count":29,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T00:00:00Z","timestamp":1777075200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Malaysia-Jordan Matching Fund","award":["MMUI\/240092"],"award-info":[{"award-number":["MMUI\/240092"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Transparent clinical decision-making remains a critical barrier to deploying deep learning in medical diagnosis. Post hoc explanation methods approximate model behaviour after training but cannot guarantee that explanations faithfully reflect the underlying reasoning. This study proposes a Self-Explaining Neural Network (SENN) for Parkinson\u2019s Disease (PD) screening via Ground Reaction Force (GRF) gait analysis, enforcing intrinsic interpretability through learnable basis concepts and input-dependent relevance scores computed jointly with the prediction. The architecture combines a four-block residual CNN backbone with stochastic depth regularisation, a 16-concept encoder with diversity and stability constraints, and temperature-scaled probability calibration for reliable clinical operating points. Evaluated on the PhysioNet Gait in Parkinson\u2019s Disease dataset (306 subjects, 16 GRF sensors per foot), SENN achieves a subject-level ROC-AUC of 0.916 [95% CI: 0.867\u20130.964], sensitivity of 0.913 [0.862\u20130.963], specificity of 0.671 [0.485\u20130.858], and Average Precision of 0.942 [0.918\u20130.967], reported across five independent random seeds. Comparative evaluation against four deep learning baselines\u2014CNN-Residual, BiLSTM, CNN-LSTM, and CNN-Attention\u2014confirms that the interpretability constraints impose no statistically significant reduction in discriminative performance, with all pairwise ROC-AUC confidence intervals overlapping. Concept-level analysis reveals that the three most discriminative concepts correspond to disrupted midfoot loading patterns, increased step-length variability, and bilateral cadence asymmetry\u2014all established biomechanical hallmarks of parkinsonian gait\u2014providing clinically grounded, patient-specific explanations without post hoc approximation. These findings demonstrate that rigorous intrinsic interpretability and competitive predictive accuracy are simultaneously achievable in deep gait analysis, supporting the clinical adoption of transparent diagnostic AI.<\/jats:p>","DOI":"10.3390\/s26092671","type":"journal-article","created":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T08:12:31Z","timestamp":1777363951000},"page":"2671","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Self-Explaining Neural Networks for Transparent Parkinson\u2019s Disease Screening"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9010-6989","authenticated-orcid":false,"given":"Mahmoud E.","family":"Farfoura","sequence":"first","affiliation":[{"name":"Cybersecurity Department, Al-Zaytoonah University of Jordan, Amman 11733, Jordan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmad A. A.","family":"Alkhatib","sequence":"additional","affiliation":[{"name":"Cybersecurity Department, Al-Zaytoonah University of Jordan, Amman 11733, Jordan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0901-3831","authenticated-orcid":false,"given":"Tee","family":"Connie","sequence":"additional","affiliation":[{"name":"Faculty of Information Science & Technology, Multimedia University, Jalan Ayer Keroh Lama, Melaka 75450, Malaysia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,25]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"7102","DOI":"10.1007\/s00415-024-12687-5","article-title":"Definition and diagnosis of Parkinson\u2019s disease: Guideline \u2018Parkinson\u2019s disease\u2019 of the German Society of Neurology","volume":"271","author":"Hopfner","year":"2024","journal-title":"J. Neurol."},{"key":"ref_2","first-page":"113","article-title":"Intelligent diagnostic prediction and classification system for Parkinson\u2019s disease by incorporating sperm swarm optimization (SSO) and density-based feature selection methods","volume":"15","author":"Shehadeh","year":"2023","journal-title":"Int. J. Adv. Soft Comput. Appl."},{"key":"ref_3","doi-asserted-by":"crossref","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":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Athamneh, A.A., Farfoura, M.E., and Rosiyadi, D. (2025). A CatBoost Predictive Model for Parkinson\u2019s Disease Early Detection. Proceedings of the 2025 1st International Conference on Computational Intelligence Approaches and Applications (ICCIAA), Amman, Jordan, 28\u201330 April 2025, IEEE.","DOI":"10.1109\/ICCIAA65327.2025.11013735"},{"key":"ref_5","unstructured":"Alvarez-Melis, D., and Jaakkola, T.S. (2018, January 2\u20138). Towards robust interpretability with self-explaining neural networks. Proceedings of the 31st Conference on Neural Information Processing Systems (NeurIPS), Montr\u00e9al, QC, Canada."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Huang, T., Li, M., and Huang, J. (2023). Recent trends in wearable device used to detect freezing of gait and falls in people with Parkinson\u2019s disease: A systematic review. Front. Aging Neurosci., 15.","DOI":"10.3389\/fnagi.2023.1119956"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Wang, Y., Beuving, F., Nonnekes, J., Cohen, M.X., Long, X., Aarts, R.M., and van Wezel, R. (2020). Freezing of gait detection in Parkinson\u2019s disease via multimodal analysis of EEG and accelerometer signals. Proceedings of the 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), IEEE.","DOI":"10.1109\/EMBC44109.2020.9175288"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"22547","DOI":"10.1038\/s41598-022-26644-7","article-title":"Detection of Parkinson\u2019s disease from EEG signals using discrete wavelet transform, different entropy measures, and machine learning techniques","volume":"12","author":"Aljalal","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Sato, K., Nagashima, Y., Mano, T., Iwata, A., and Toda, T. (2019). Quantifying normal and parkinsonian gait features from home movies: Practical application of a deep learning\u2013based 2D pose estimator. PLoS ONE, 14.","DOI":"10.1101\/782367"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"111278","DOI":"10.1016\/j.jbiomech.2022.111278","article-title":"Concurrent validity of artificial intelligence-based markerless motion capture for over-ground gait analysis: A study of spatiotemporal parameters","volume":"143","author":"Ripic","year":"2022","journal-title":"J. Biomech."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Rupprechter, S., Morinan, G., Peng, Y., Foltynie, T., Sibley, K., Weil, R.S., Leyland, L.-A., Baig, F., Morgante, F., and Gilron, R. (2021). A clinically interpretable computer-vision based method for quantifying gait in Parkinson\u2019s disease. Sensors, 21.","DOI":"10.3390\/s21165437"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2432","DOI":"10.1038\/s41598-019-38623-6","article-title":"Translational methods to detect asymmetries in temporal and spatial walking metrics in parkinsonian mouse models and human subjects with Parkinson\u2019s disease","volume":"9","author":"Broom","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Yin, Z., Jiang, Y., Zheng, J., and Yu, H. (2023). STJA-GCN: A Multi-Branch Spatial\u2013Temporal Joint Attention Graph Convolutional Network for Abnormal Gait Recognition. Appl. Sci., 13.","DOI":"10.3390\/app13074205"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"23826","DOI":"10.1109\/ACCESS.2021.3056880","article-title":"Multi-Model Long Short-Term Memory Network for Gait Recognition Using Window-Based Data Segment","volume":"9","author":"Tran","year":"2021","journal-title":"IEEE Access"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2912","DOI":"10.1109\/TNSRE.2023.3291359","article-title":"Video-Based Quantification of Gait Impairments in Parkinson\u2019s Disease Using Skeleton-Silhouette Fusion Convolution Network","volume":"31","author":"Zeng","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Yu, Z., Zhao, J., Zhou, X., Liu, K., and Yan, Y. (2021). Gait Phase Detection Based on a Foot-Mounted Inertial Sensor Using Long Short-Term Memory Enhanced by Hidden Markov Model. Proceedings of the 26th International Conference on Automation and Computing (ICAC), IEEE.","DOI":"10.23919\/ICAC50006.2021.9594161"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"133","DOI":"10.3991\/ijoe.v22i01.58513","article-title":"Parkinson\u2019s disease detection through offline handwriting analysis: A CNN-based approach","volume":"22","author":"Bensefia","year":"2026","journal-title":"Int. J. Online Biomed. Eng. (iJOE)"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Khedimi, M., Zhang, T., Dehmani, C., Zhao, X., and Geng, Y. (2025). A Unified Deep Learning Ensemble Framework for Voice-Based Parkinson\u2019s Disease Detection and Motor Severity Prediction. Bioengineering, 12.","DOI":"10.3390\/bioengineering12070699"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Rangel-Cascajosa, C., Luna-Perej\u00f3n, F., Vicente-Diaz, S., and Dom\u00ednguez-Morales, M. (2025). Gait-Based Parkinson\u2019s Disease Detection Using Recurrent Neural Networks for Wearable Systems. Big Data Cogn. Comput., 9.","DOI":"10.3390\/bdcc9070183"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Nayan, N.M., Rana, A.M., Islam, M.M., Uddin, J., Yasmin, T., and Uddin, J. (2025). An interpretable and balanced machine learning framework for Parkinson\u2019s disease prediction using feature engineering and explainable AI. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0333418"},{"key":"ref_21","first-page":"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":"ref_22","doi-asserted-by":"crossref","first-page":"104668","DOI":"10.1016\/j.pdpdt.2025.104668","article-title":"Decoding Parkinson\u2019s Diagnosis: An OCT-Based Explainable AI with SHAP\/LIME Transparency from the Persian Cohort Study","volume":"54","author":"Shoeibi","year":"2025","journal-title":"Photodiagnosis Photodyn. Ther."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"6046","DOI":"10.1609\/aaai.v36i6.20551","article-title":"Entropy-Based Logic Explanations of Neural Networks","volume":"Volume 36","author":"Barbiero","year":"2022","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"103822","DOI":"10.1016\/j.artint.2022.103822","article-title":"Logic Explained Networks","volume":"314","author":"Ciravegna","year":"2023","journal-title":"Artif. Intell."},{"key":"ref_25","unstructured":"Chen, C., Li, O., Tao, D., Barnett, A., Rudin, C., and Su, J. (2019, January 8\u201314). This Looks Like That: Deep Learning for Interpretable Image Recognition. Proceedings of the Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, Vancouver, BC, Canada."},{"key":"ref_26","doi-asserted-by":"crossref","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":"ref_27","unstructured":"Guo, C., Pleiss, G., Sun, Y., and Weinberger, K.Q. On calibration of modern neural networks. International Conference on Machine Learning (ICML), PMLR."},{"key":"ref_28","unstructured":"Loshchilov, I., and Hutter, F. (2019, January 6\u20139). Decoupled Weight Decay Regularization. Proceedings of the International Conference on Learning Representations (ICLR), New Orleans, LA, USA. Available online: https:\/\/openreview.net\/forum?id=Bkg6RiCqY7."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"AlZu\u2019bi, S., Elbes, M., Mughaid, A., Bdair, N., Abualigah, L., Forestiero, A., and Zitar, R.A. (2023). Diabetes Monitoring System in Smart Health Cities Based on Big Data Intelligence. Future Internet, 15.","DOI":"10.3390\/fi15020085"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/9\/2671\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T08:21:15Z","timestamp":1777364475000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/9\/2671"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,25]]},"references-count":29,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["s26092671"],"URL":"https:\/\/doi.org\/10.3390\/s26092671","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,25]]}}}