{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T05:17:20Z","timestamp":1787894240555,"version":"build-2784847793"},"reference-count":38,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T00:00:00Z","timestamp":1787702400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Bioengineering"],"abstract":"<jats:p>Background: Hypertensive retinopathy (HR) is an early marker of hypertension-related end-organ damage, and multi-grade classification is limited by overlap between the early stages. No prior study has reported edge-deployable HR grading from optical coherence tomography (OCT). We assess patient-level\u2013validated four-grade HR classification from en-face OCT with on-device inference. Methods: MobileViT-XXS and EfficientNetV2-B0 were evaluated on 478 en-face OCT images from 221 patients. To avoid leakage, all partitions were patient-level: stratified group 5-fold cross-validation with a patient-disjoint test set. Each model was evaluated as a 5-fold soft-voting ensemble, with inference profiled on an NVIDIA Jetson Orin Nano. Accuracy, weighted F1, quadratic-weighted kappa, and ROC\u2013AUC were reported; the models were compared by McNemar\u2019s test, and clinical utility by referable-HR (Grade \u2265 2) triage. Results: MobileViT-XXS achieved 84.4% accuracy, ROC\u2013AUC 0.944, and quadratic-weighted kappa 0.916, significantly outperforming EfficientNetV2-B0 (McNemar p = 0.013). Mild-grade overlap limited four-class accuracy, but referable-HR triage reached 95.6% accuracy, 95.2% sensitivity, and 95.8% negative predictive value. MobileViT-XXS required only 3.83 MB versus 23.70 MB, and on-device TensorRT FP16 five-fold ensemble inference achieved 10.4 ms (96 FPS) at 8.9 W. Conclusions: Lightweight models enable feasible, power-efficient point-of-care HR triage from en-face OCT; broader multi-center validation is warranted.<\/jats:p>","DOI":"10.3390\/bioengineering13090984","type":"journal-article","created":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T15:04:32Z","timestamp":1787756672000},"page":"984","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Edge-Deployable Lightweight Deep Learning for Hypertensive Retinopathy Grading from En-Face OCT: A Patient-Level Feasibility Study"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2808-1861","authenticated-orcid":false,"given":"S\u00fcleyman Bur\u00e7in","family":"\u015e\u00fcyun","sequence":"first","affiliation":[{"name":"Digital Transformation Office, The Rectorate of Sinop University, Sinop University, 57000 Sinop, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0562-4931","authenticated-orcid":false,"given":"Mustafa","family":"Yurdakul","sequence":"additional","affiliation":[{"name":"Computer Engineering Department, K\u0131r\u0131kkale University, 71450 K\u0131r\u0131kkale, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2433-246X","authenticated-orcid":false,"given":"\u015eakir","family":"Ta\u015fdemir","sequence":"additional","affiliation":[{"name":"Digital Transformation Office, The Rectorate of Sinop University, Sinop University, 57000 Sinop, T\u00fcrkiye"},{"name":"Computer Engineering Department, Sel\u00e7uk University, 42250 Konya, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Serkan","family":"Bili\u015f","sequence":"additional","affiliation":[{"name":"Eye Diseases Department, Bat\u0131g\u00f6z Medical Group Hospital, 35200 \u0130zmir, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,26]]},"reference":[{"key":"ref_1","unstructured":"World Health Organization (2023). Global Report on Hypertension: The Race Against a Silent Killer, World Health Organization."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1038\/s41581-019-0244-2","article-title":"The global epidemiology of hypertension","volume":"16","author":"Mills","year":"2020","journal-title":"Nat. Rev. Nephrol."},{"key":"ref_3","first-page":"6424","article-title":"A literature review of hypertensive retinopathy: Systemic correlations and new technologies","volume":"26","author":"Aiello","year":"2022","journal-title":"Eur. Rev. Med. Pharmacol. Sci."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"332","DOI":"10.1097\/00000441-193903000-00006","article-title":"Some different types of essential hypertension: Their course and prognosis","volume":"197","author":"Keith","year":"1939","journal-title":"Am. J. Med. Sci."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"2310","DOI":"10.1056\/NEJMra032865","article-title":"Hypertensive retinopathy","volume":"351","author":"Wong","year":"2004","journal-title":"N. Engl. J. Med."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1097\/APO.0000000000000509","article-title":"The new era of retinal imaging in hypertensive patients","volume":"11","author":"Tan","year":"2022","journal-title":"Asia-Pac. J. Ophthalmol."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Abbas, Q., Daadaa, Y., Rashid, U., Sajid, M.Z., and Ibrahim, M.E. (2023). HDR-EfficientNet: A classification of hypertensive and diabetic retinopathy using optimize EfficientNet architecture. Diagnostics, 13.","DOI":"10.3390\/diagnostics13203236"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1016\/j.procs.2017.10.066","article-title":"The classification of hypertensive retinopathy using convolutional neural network","volume":"116","author":"Triwijoyo","year":"2017","journal-title":"Procedia Comput. Sci."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"20220108","DOI":"10.1259\/dmfr.20220108","article-title":"Comparison of five convolutional neural networks for predicting osteoporosis based on mandibular cortical index on panoramic radiographs","volume":"51","author":"Tassoker","year":"2022","journal-title":"Dentomaxillofacial Radiol."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Qureshi, I., Abbas, Q., Yan, J., Hussain, A., Shaheed, K., and Baig, A.R. (2022). Computer-aided detection of hypertensive retinopathy using depth-wise separable CNN. Appl. Sci., 12.","DOI":"10.3390\/app122312086"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Bhimavarapu, U., Chintalapudi, N., and Battineni, G. (2024). Automatic detection and classification of hypertensive retinopathy with improved convolution neural network and improved SVM. Bioengineering, 11.","DOI":"10.3390\/bioengineering11010056"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"\u015e\u00fcyun, S.B., Yurdakul, M., Ta\u015fdemir, \u015e., and Bili\u015f, S. (2025). Triple-stream deep feature selection with metaheuristic optimization and machine learning for multi-stage hypertensive retinopathy diagnosis. Appl. Sci., 15.","DOI":"10.3390\/app15126485"},{"key":"ref_13","unstructured":"Mehta, S., and Rastegari, M. (2021). MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer. arXiv."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"117009","DOI":"10.1016\/j.eswa.2022.117009","article-title":"Detecting retinal vasculature as a key biomarker for deep Learning-based intelligent screening and analysis of diabetic and hypertensive retinopathy","volume":"200","author":"Arsalan","year":"2022","journal-title":"Expert Syst. Appl."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"108585","DOI":"10.1016\/j.cmpb.2025.108585","article-title":"Severity grading of hypertensive retinopathy using hybrid deep learning architecture","volume":"261","author":"Suman","year":"2025","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"e70159","DOI":"10.1002\/ima.70159","article-title":"MaxGlaViT: A Novel Lightweight Vision Transformer-Based Approach for Early Diagnosis of Glaucoma Stages From Fundus Images","volume":"35","author":"Yurdakul","year":"2025","journal-title":"Int. J. Imaging Syst. Technol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1136\/bmj.331.7508.73","article-title":"Value of routine funduscopy in patients with hypertension: Systematic review","volume":"331","author":"Hulsman","year":"2005","journal-title":"BMJ"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1103","DOI":"10.1016\/S0140-6736(89)92384-2","article-title":"Usefulness of ophthalmoscopy in mild to moderate hypertension","volume":"333","author":"Dimmitt","year":"1989","journal-title":"Lancet"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Muchuchuti, S., and Viriri, S. (2023). Retinal disease detection using deep learning techniques: A comprehensive review. J. Imaging, 9.","DOI":"10.3390\/jimaging9040084"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"107627","DOI":"10.1016\/j.cmpb.2023.107627","article-title":"Computer-aided diagnostic system for hypertensive retinopathy: A review","volume":"240","author":"Suman","year":"2023","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1007\/s10278-024-01218-3","article-title":"Deep convolutional neural network for automated staging of periodontal bone loss severity on bite-wing radiographs: An Eigen-CAM explainability mapping approach","volume":"38","author":"Erturk","year":"2025","journal-title":"J. Imaging Inform. Med."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Du, K., Zheng, Y., and Wang, Z. (2026). Artificial intelligence-integrated multimodal retinal imaging for early detection and risk stratification of systemic vascular and neurodegenerative diseases. Front. Neurol., 17.","DOI":"10.3389\/fneur.2026.1885303"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"9","DOI":"10.21037\/aes-23-10","article-title":"Advances in teleophthalmology and artificial intelligence for diabetic retinopathy screening: A narrative review","volume":"9","author":"Chokshi","year":"2024","journal-title":"Ann. Eye Sci."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1038\/s41746-022-00592-y","article-title":"Machine learning for medical imaging: Methodological failures and recommendations for the future","volume":"5","author":"Varoquaux","year":"2022","journal-title":"NPJ Digit. Med."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"14490","DOI":"10.1038\/s41598-021-93651-5","article-title":"Impact of train\/test sample regimen on performance estimate stability of machine learning in cardiovascular imaging","volume":"11","author":"Singh","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_26","unstructured":"Tan, M., and Le, Q. (2021, January 18\u201324). Efficientnetv2: Smaller models and faster training. Proceedings of the 38th International Conference on Machine Learnin, Virtual."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"712","DOI":"10.1007\/s11760-025-04321-2","article-title":"BC-YOLO: MBConv-ECA based YOLO framework for blood cell detection","volume":"19","author":"Yurdakul","year":"2025","journal-title":"Signal Image Video Process."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"4027","DOI":"10.1007\/s12596-025-02793-6","article-title":"An enhanced lightweight transformer-based framework for accurate retinal disease classification from OCT images","volume":"55","author":"Mani","year":"2026","journal-title":"J. Opt."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"110164","DOI":"10.1016\/j.compbiomed.2025.110164","article-title":"Effivit: Hybrid cnn-transformer for retinal imaging","volume":"191","author":"Ashoka","year":"2025","journal-title":"Comput. Biol. Med."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"2679","DOI":"10.1007\/s00784-022-04839-6","article-title":"Detection of pulpal calcifications on bite-wing radiographs using deep learning","volume":"27","author":"Yuce","year":"2023","journal-title":"Clin. Oral Investig."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"30554","DOI":"10.1038\/s41598-024-81132-4","article-title":"Ensemble deep learning and EfficientNet for accurate diagnosis of diabetic retinopathy","volume":"14","author":"Arora","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Yurdakul, M., and Ta\u015fdemir, \u015e. (2023, January 27\u201328). Brain tumor detection with ensemble of convolutional neural networks and vision transformer. Proceedings of the 2023 2nd International Engineering Conference on Electrical, Energy, and Artificial Intelligence (EICEEAI), Zarqa, Jordan.","DOI":"10.1109\/EICEEAI60672.2023.10590129"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Moshkin, A., Fedorov, M., Arlazarov, V., Gribova, V., Nazarenko, A., Repin, D., Klevtsova, O., and Romanov, A. (2026). Localization in Medical Imaging: A Unified AI Approach for Ovaries, Follicles, and Vertebral Arteries. Algorithms, 19.","DOI":"10.3390\/a19070523"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"e003424","DOI":"10.1136\/bmjdrc-2023-003424","article-title":"Artificial intelligence for diabetic retinopathy in low-income and middle-income countries: A scoping review","volume":"11","author":"Cleland","year":"2023","journal-title":"BMJ Open Diabetes Res. Care"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"715","DOI":"10.1007\/s40846-023-00831-x","article-title":"Fully Automated Detection of Osteoporosis Stage on Panoramic Radiographs Using YOLOv5 Deep Learning Model and Designing a Graphical User Interface","volume":"43","author":"Tassoker","year":"2023","journal-title":"J. Med. Biol. Eng."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"108755","DOI":"10.1016\/j.compbiomed.2024.108755","article-title":"AI-Driven localization of all impacted teeth and prediction of winter angulation for third molars on panoramic radiographs: Clinical user interface design","volume":"178","author":"Zirek","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1007\/s10341-025-01327-5","article-title":"Webserver-Based mobile application for Multi-class chestnut (Castanea sativa) classification using deep features and attention mechanisms","volume":"67","author":"Yurdakul","year":"2025","journal-title":"Appl. Fruit Sci."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"106617","DOI":"10.1016\/j.bspc.2024.106617","article-title":"ABC-based weighted voting deep ensemble learning model for multiple eye disease detection","volume":"96","author":"Uyar","year":"2024","journal-title":"Biomed. Signal Process. Control"}],"container-title":["Bioengineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/9\/984\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T04:17:11Z","timestamp":1787890631000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/9\/984"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,26]]},"references-count":38,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["bioengineering13090984"],"URL":"https:\/\/doi.org\/10.3390\/bioengineering13090984","relation":{},"ISSN":["2306-5354"],"issn-type":[{"value":"2306-5354","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,26]]}}}