{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T18:10:29Z","timestamp":1790100629815,"version":"4.0.1"},"reference-count":32,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Parkinson\u2019s disease (PD) is a neurodegenerative disorder fundamentally characterized by motor impairments and diminished facial expressivity. This study investigates the efficacy of feature engineering approaches applied to facial video data captured via smartphone frontal cameras to predict PD presence, disease severity, and motor function scores. Within a cohort of 40 participants, a combination of three predictive models, four facial movement scenarios, and five frame selection strategies was rigorously evaluated. Regarding PD classification, the XGBoost model achieved maximum performance using the smiling video scenario with frame skipping combined with PCA retaining 95% of the variance, yielding an accuracy of 0.77, a precision of 0.75, a recall of 0.86, and an F1-score of 0.79. For Hoehn &amp; Yahr severity estimation, the Random Forest model implemented with the smiling video, frame skipping, and no PCA attained the best result, with the lowest RMSE of 1.30 and a MAPE of 38.36%. Furthermore, in The Movement Disorder Society-sponsored revision of the Unified Parkinson\u2019s Disease Rating Scale (MDS-UPDRS) Part III motor score estimation, the XGBoost model outperformed other configurations using the smiling video with 3-frame averaging and PCA retaining 95% of the variance, achieving an RMSE of 19.34 and a MAPE of 45.33%. Crucially, the Explainable AI analysis identified that lip and mouth corner movements were the most critical features for both PD classification and Hoehn &amp; Yahr estimation. In contrast, eyebrow and eye movements were the dominant features for MDS-UPDRS Part III motor score estimation. These findings indicate that facial dynamics carry information relevant to PD assessment and support further development of non-invasive, accessible screening tools, although the accuracy achieved here is not yet sufficient to substitute for clinical evaluation.<\/jats:p>","DOI":"10.1371\/journal.pone.0357902","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T17:34:08Z","timestamp":1790098448000},"page":"e0357902","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Parkinson\u2019s disease assessment with selfie video via feature engineering and explainable AI"],"prefix":"10.1371","volume":"21","author":[{"given":"Sitthatka","family":"Jaratsaeng","sequence":"first","affiliation":[{"name":"Khon Kaen University","place":["Khon Kaen, Thailand"],"department":["Visual Intelligence Laboratory, Faculty of Science"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4230-5641","authenticated-orcid":true,"given":"Anchalee","family":"Techasen","sequence":"additional","affiliation":[{"name":"Khon Kaen University","place":["Khon Kaen, Thailand"],"department":["Faculty of Associated Medical Sciences"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Narongrit","family":"Kasemsap","sequence":"additional","affiliation":[{"name":"Khon Kaen University","place":["Khon Kaen, Thailand"],"department":["Department of Medicine, Faculty of Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0965-680X","authenticated-orcid":true,"given":"Thanapong","family":"Intharah","sequence":"additional","affiliation":[{"name":"Khon Kaen University","place":["Khon Kaen, Thailand"],"department":["Visual Intelligence Laboratory, Faculty of Science"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"issue":"9996","key":"pone.0357902.ref001","doi-asserted-by":"crossref","first-page":"896","DOI":"10.1016\/S0140-6736(14)61393-3","article-title":"Parkinson\u2019s disease","volume":"386","author":"LV Kalia","year":"2015","journal-title":"Lancet"},{"key":"pone.0357902.ref002","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1136\/jnnp.52.Suppl.6","article-title":"Aspects of the history of Parkinson\u2019s disease","author":"JM Pearce","year":"1989","journal-title":"J Neurol Neurosurg Psychiatry"},{"issue":"3","key":"pone.0357902.ref003","first-page":"381","article-title":"Monoamine precursors and analogues","volume":"1","author":"A Carlsson","year":"1975","journal-title":"Pharmacol Ther B"},{"issue":"1","key":"pone.0357902.ref004","doi-asserted-by":"crossref","DOI":"10.1101\/cshperspect.a008888","article-title":"Genetics of Parkinson\u2019s disease","volume":"2","author":"C Klein","year":"2012","journal-title":"Cold Spring Harbor Perspectives in Medicine"},{"issue":"1","key":"pone.0357902.ref005","doi-asserted-by":"crossref","DOI":"10.1101\/cshperspect.a008862","article-title":"The history of Parkinson\u2019s disease: early clinical descriptions and neurological therapies","volume":"1","author":"CG Goetz","year":"2011","journal-title":"Cold Spring Harb Perspect Med"},{"key":"pone.0357902.ref006","article-title":"Neuronal pathology in Parkinson\u2019s disease","volume":"251","author":"JB Schulz","year":"2004","journal-title":"Journal of Neurology"},{"issue":"12","key":"pone.0357902.ref007","doi-asserted-by":"crossref","first-page":"2422","DOI":"10.1111\/ene.14452","article-title":"Hypomimia in Parkinson\u2019s disease: an axial sign responsive to levodopa","volume":"27","author":"L Ricciardi","year":"2020","journal-title":"Eur J Neurol"},{"issue":"5","key":"pone.0357902.ref008","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1212\/WNL.17.5.427","article-title":"Parkinsonism: onset, progression and mortality","volume":"17","author":"MM Hoehn","year":"1967","journal-title":"Neurology"},{"issue":"15","key":"pone.0357902.ref009","doi-asserted-by":"crossref","first-page":"2129","DOI":"10.1002\/mds.22340","article-title":"Movement Disorder Society-sponsored revision of the Unified Parkinson\u2019s Disease Rating Scale (MDS-UPDRS","volume":"23","author":"CG Goetz","year":"2008","journal-title":"). Movement Disorders"},{"key":"pone.0357902.ref010","doi-asserted-by":"crossref","first-page":"765765","DOI":"10.3389\/fnins.2021.765765","article-title":"Hoehn and Yahr Stage and Striatal Dat-SPECT Uptake Are Predictors of Parkinson\u2019s Disease Motor Progression","volume":"15","author":"PASADENA Investigators, Prasinezumab Study Group","year":"2021","journal-title":"Front Neurosci"},{"key":"pone.0357902.ref011","doi-asserted-by":"crossref","unstructured":"Jaratsaeng S, Techasen A, Kasemsap N, Intharah T. PDFace: Parkinson\u2019s Disease Classification Via Movement Analysis of Video From Mobile Camera. In: 2026 18th International Conference on Knowledge and Smart Technology (KST), 2026. 59\u201364. https:\/\/doi.org\/10.1109\/kst67832.2026.11432360","DOI":"10.1109\/KST67832.2026.11432360"},{"key":"pone.0357902.ref012","doi-asserted-by":"crossref","unstructured":"Baltru\u0161aitis T, Zadeh A, Lim YC, Morency LP. OpenFace 2.0: Facial Behavior Analysis Toolkit. In: 2018 13th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2018); 2018. p. 59\u201366. https:\/\/doi.org\/10.1109\/FG.2018.00019","DOI":"10.1109\/FG.2018.00019"},{"key":"pone.0357902.ref013","first-page":"1755","article-title":"Dlib-ml: A Machine Learning Toolkit","volume":"10","author":"D King","year":"2009","journal-title":"Journal of Machine Learning Research"},{"key":"pone.0357902.ref014","doi-asserted-by":"crossref","unstructured":"Baltrusaitis T, Robinson P, Morency L-P. OpenFace: An open source facial behavior analysis toolkit. In: 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), 2016. 1\u201310. https:\/\/doi.org\/10.1109\/wacv.2016.7477553","DOI":"10.1109\/WACV.2016.7477553"},{"key":"pone.0357902.ref015","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1016\/j.jneumeth.2017.02.006","article-title":"Analysis of facial expressions in parkinson\u2019s disease through video-based automatic methods","volume":"281","author":"A Bandini","year":"2017","journal-title":"J Neurosci Methods"},{"key":"pone.0357902.ref016","doi-asserted-by":"crossref","unstructured":"Rajnoha M, Mekyska J, Burget R, Eliasova I, Kostalova M, Rektorova I. Towards Identification of Hypomimia in Parkinson\u2019s Disease Based on Face Recognition Methods. In: 2018 10th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), 2018. 1\u20134. https:\/\/doi.org\/10.1109\/icumt.2018.8631249","DOI":"10.1109\/ICUMT.2018.8631249"},{"key":"pone.0357902.ref017","first-page":"249","article-title":"Visual Detection of Parkinson\u2019s Disease via Facial Features Recognition","volume-title":"Lecture Notes in Electrical Engineering","author":"X Hou","year":"2021"},{"key":"pone.0357902.ref018","doi-asserted-by":"crossref","first-page":"627199","DOI":"10.3389\/fnagi.2020.627199","article-title":"Motor progression in early-stage Parkinson\u2019s disease: Prediction using regression and machine learning models","volume":"12","author":"LY Ma","year":"2021","journal-title":"Frontiers in Aging Neuroscience"},{"key":"pone.0357902.ref019","article-title":"Unmasking Parkinson\u2019s Disease with Smile: An AI-enabled Screening Framework","author":"T Adnan","year":"2023","journal-title":"arXiv preprint"},{"key":"pone.0357902.ref020","unstructured":"Duan S, Dadashzadeh A, Whone A, Mirmehdi M. QAFE-Net: Quality Assessment of Facial Expressions with Landmark Heatmaps. In: Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV); 2024. p. 6308\u201317."},{"key":"pone.0357902.ref021","article-title":"Accessible, At-Home Detection of Parkinson\u2019s Disease via Multi-task Video Analysis","author":"MS Islam","year":"2024","journal-title":"arXiv preprint"},{"issue":"20","key":"pone.0357902.ref022","doi-asserted-by":"crossref","first-page":"6373","DOI":"10.3390\/s25206373","article-title":"AI Video Analysis in Parkinson\u2019s Disease: A Systematic Review of the Most Accurate Computer Vision Tools for Diagnosis, Symptom Monitoring, and Therapy Management","volume":"25","author":"L di Biase","year":"2025","journal-title":"Sensors (Basel)"},{"issue":"8","key":"pone.0357902.ref023","doi-asserted-by":"crossref","first-page":"782","DOI":"10.3390\/brainsci15080782","article-title":"Predicting Cognitive Decline in Parkinson\u2019s Disease Using Artificial Neural Networks: An Explainable AI Approach","volume":"15","author":"L Colautti","year":"2025","journal-title":"Brain Sciences"},{"key":"pone.0357902.ref024","volume-title":"Principal Component Analysis","author":"IT Jolliffe","year":"2002","edition":"2"},{"issue":"3","key":"pone.0357902.ref025","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-Vector Networks","volume":"20","author":"C Cortes","year":"1995","journal-title":"Machine Learning"},{"issue":"1","key":"pone.0357902.ref026","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random Forests","volume":"45","author":"L Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"pone.0357902.ref027","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2016. p. 785\u201394.","DOI":"10.1145\/2939672.2939785"},{"key":"pone.0357902.ref028","unstructured":"Kohavi R. A study of cross-validation and bootstrap for accuracy estimation and model selection. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence; 1995. p. 1137\u201345."},{"key":"pone.0357902.ref029","volume-title":"Parkinson\u2019s disease assessment using mobile phone camera","author":"S Jaratsaeng","year":"2026"},{"key":"pone.0357902.ref030","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research"},{"issue":"9","key":"pone.0357902.ref031","doi-asserted-by":"crossref","first-page":"1020","DOI":"10.1002\/mds.20213","article-title":"Movement Disorder Society Task Force report on the Hoehn and Yahr staging scale: status and recommendations","volume":"19","author":"CG Goetz","year":"2004","journal-title":"Movement Disorders"},{"issue":"12","key":"pone.0357902.ref032","doi-asserted-by":"crossref","first-page":"1421","DOI":"10.1016\/j.parkreldis.2015.10.006","article-title":"Minimal clinically important difference on the Motor Examination part of MDS-UPDRS","volume":"21","author":"K Horv\u00e1th","year":"2015","journal-title":"Parkinsonism Relat Disord"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0357902","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T17:34:19Z","timestamp":1790098459000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0357902"}},"subtitle":[],"editor":[{"given":"Ali","family":"Al-Naji","sequence":"first","affiliation":[{"name":"Middle Technical University","place":["IRAQ"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,22]]},"references-count":32,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,22]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0357902","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,22]]}}}