{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T06:05:24Z","timestamp":1790661924744,"version":"4.1.0"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"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. Digit. Health"],"abstract":"<jats:p>Accurate and timely detection of sickle red blood cells has a significant impact on the patients\u2019 treatment. Developing a computer vision system that is both lightweight and resource-efficient is important for enabling real-time edge-based applications in resource-constrained healthcare environments. This study proposes a hybrid sickle cell detection framework that uses Bi-lightweight Knowledge Distillation (KD)-guided CNNs, combining two lightweight student models from different teacher networks with Machine Learning (ML) classifiers to identify sickle cells with accuracy and efficiency. Using depthwise separable convolution layers, lightweight models are created that drastically lower model complexity. These lightweight models are improved by extracting knowledge from top-performing teacher networks, such as MobileNetV2 and DenseNet121. Using this Knowledge Distillation learning technique, lightweight models can enhance their representational capabilities without adding to their computational complexity. To improve decision-making performance, the Bi-distilled student models are combined with conventional machine learning KNN classifiers to create a hybrid ensemble that combines the advantages of ML-based classification with dual KD-guided CNN-based feature extraction. The Sickle Cell Disease Dataset is used to train and validate the proposed system. The learned student models are then converted into TensorFlow Lite (TFLite) format using Float32 quantization to further reduce computational complexity. In combination with dual KD-guided CNN models and a KNN classifier, the proposed hybrid model achieves a high classification accuracy of 95.45% and an F1-score of 95.02% while being small at only 3.96 MB. Deploying and evaluating the framework on a resource-constrained Raspberry Pi 4 Model B demonstrate its potential for real-time, edge-based, and point-of-care applications.<\/jats:p>","DOI":"10.3389\/fdgth.2026.1938284","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:51:02Z","timestamp":1790661062000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Hybrid multi-model framework with knowledge distillation for precise and efficient sickle cell detection"],"prefix":"10.3389","volume":"8","author":[{"given":"Metta Manikanta Karthikeya","family":"Reddy","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cherukuri Krishna","family":"Lathvik","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arumuga Arun","family":"R","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Karthik","family":"K","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"934","DOI":"10.3390\/cells13110934","article-title":"End organ affection in sickle cell disease","volume":"13","author":"Bathla","year":"2024","journal-title":"Cells"},{"key":"B2","first-page":"1","article-title":"Red blood cell transfusions may have the strongest analgesic effect during acute painful crises in sickle cell diseases","volume":"13","author":"Helvaci","year":"2024","journal-title":"Ann Clin Med Case Rep"},{"key":"B3","doi-asserted-by":"publisher","first-page":"546","DOI":"10.3324\/haematol.2024.285066","article-title":"Infections in sickle cell disease","volume":"110","author":"Scourfield","year":"2024","journal-title":"Haematologica"},{"key":"B4","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1093\/ajcp\/aqae042","article-title":"Concentration of voxelotor in sickle cell disease can be estimated using electrophoresis and high-performance liquid chromatography","volume":"162","author":"Curtis","year":"2024","journal-title":"Am J Clin Pathol"},{"key":"B5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/ICCA62237.2024.10927806","article-title":"Advancements in machine learning and deep learning techniques for sickle cell disease detection using digital morphology","volume-title":"2024 International Conference on Computer and Applications (ICCA)","author":"Abdul","year":"2024"},{"key":"B6","author":"Cardoso","year":"2024"},{"key":"B7","author":"Jones","year":"2024"},{"key":"B8","doi-asserted-by":"crossref","DOI":"10.36227\/techrxiv.173161086.63114554\/v1","volume-title":"An Image-based Sickle Cell Detection Method","author":"Tushabe","year":"2024"},{"key":"B9","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1109\/ICASI60819.2024.10547902","article-title":"Automatic detection of sickle cell disease, elliptocytosis and schizocytosis","volume-title":"2024 10th International Conference on Applied System Innovation (ICASI)","author":"Nigan","year":"2024"},{"key":"B10","doi-asserted-by":"publisher","first-page":"e0313315","DOI":"10.1371\/journal.pone.0313315","article-title":"Exploring machine learning algorithms in sickle cell disease patient data: a systematic review","volume":"19","author":"Machado","year":"2024","journal-title":"PLoS One"},{"key":"B11","doi-asserted-by":"publisher","first-page":"51","DOI":"10.5220\/0012343800003651","article-title":"Machine learning-based disease severity prediction in sickle cell patients: spectroscopic insights","author":"Roy","year":"2024","journal-title":"Proceedings of the 12th International Conference on Photonics, Optics and Laser Technology (PHOTOPTICS 2024)"},{"key":"B12","doi-asserted-by":"publisher","first-page":"e22203","DOI":"10.1016\/j.heliyon.2023.e22203","article-title":"Sickle cell disease classification using deep learning","volume":"11","author":"Jennifer","year":"2023","journal-title":"Heliyon"},{"key":"B13","doi-asserted-by":"publisher","first-page":"20230179","DOI":"10.1515\/jisys-2023-0179","article-title":"Detection of sickle cell disease using deep neural networks and explainable artificial intelligence","volume":"33","author":"Goswami","year":"2024","journal-title":"J Intell Syst"},{"key":"B14","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1007\/978-3-031-67547-8_9","article-title":"Machine learning-based models for the preemptive diagnosis of sickle cell Anemia using clinical data","volume-title":"Finance and Law in the Metaverse World","author":"Olatunji","year":"2024"},{"key":"B15","doi-asserted-by":"publisher","first-page":"3741","DOI":"10.32604\/cmc.2025.061998","article-title":"Ensemble deep learning approaches in health care: a review","volume":"82","author":"Alotaibi","year":"2025","journal-title":"Comput Mater Contin"},{"key":"B16","doi-asserted-by":"publisher","first-page":"10","DOI":"10.5281\/zenodo.14591579","article-title":"Sickle cell Anemia detection using deep learning","volume":"2","author":"Subhaga","year":"2025","journal-title":"Int J Sci Res Technol"},{"key":"B17","first-page":"606","article-title":"Hybrid deep learning and machine learning approach for sickle cell disease diagnosis using ResNet and random forest classifier","volume-title":"2024 13th International Conference on System Modeling & Advancement in Research Trends (SMART)","author":"Jain","year":"2024"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1293","DOI":"10.18280\/ts.410318","article-title":"Deep learning-based red blood cell classification for sickle cell Anemia diagnosis using hybrid CNN-LSTM model","volume":"41","author":"Deo","year":"2024","journal-title":"Traitement du Signal"},{"key":"B19","first-page":"1","article-title":"A prediction of complications in sickle cell disease patients using regularized random forest","volume-title":"2024 First International Conference on Software, Systems and Information Technology (SSITCON)","author":"Alzubaidi","year":"2024"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1","DOI":"10.4108\/eetpht.10.5691","article-title":"Enhancing disease diagnosis: statistical analysis of haematological parameters in sickle cell patients, integrating predictive analytics","volume":"10","author":"Dash","year":"2024","journal-title":"EAI Endorsed Trans Pervasive Health Technol"},{"key":"B21","volume-title":"Integrating Deep Learning and Genetic Data to Detect and Predict Hereditary Transfer of Sickle Cell Anemia a ReviewProceedings of the International Conference on Innovative Computing & Communication (ICICC 2024)","author":"Pal","year":"2024"},{"key":"B22","author":"Tushabe","year":"2023"},{"key":"B23","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/IHTC61819.2024.10855129","article-title":"Sickle cell disease patient\/provider match tool","volume-title":"2024 IEEE International Humanitarian Technologies Conference (IHTC)","author":"Liang","year":"2024"},{"key":"B24","doi-asserted-by":"publisher","first-page":"15","DOI":"10.14569\/IJACSA.2024.0151202","article-title":"A framework for privacy-preserving detection of sickle blood cells using deep learning and cryptographic techniques","volume":"15","author":"Alotaibi","year":"2024","journal-title":"Int J Adv Comput Sci Appl"},{"key":"B25","first-page":"1","article-title":"Inceptionv3, ResNet50, ResNet18 and MobileNetV2 performance comparison on face recognition classification","volume":"4","author":"Ariefwan","year":"2023","journal-title":"Literasi Nusantara"},{"key":"B26","doi-asserted-by":"publisher","first-page":"863","DOI":"10.18196\/jrc.v5i3.21805","article-title":"Improving cbir techniques with deep learning approach: an ensemble method using nasnetmobile, densenet121, and vgg12","volume":"5","author":"Sadiq","year":"2024","journal-title":"JRC"},{"key":"B27","volume-title":"Enhancing Cervical Cancer Classification: Through a Hybrid Deep Learning Approach Integrating DenseNet201 and InceptionV3","author":"Sharma","year":"2025"},{"key":"B28","doi-asserted-by":"publisher","first-page":"13904","DOI":"10.1038\/s41598-025-98015-x","article-title":"Deep learning based ensemble model for accurate tomato leaf disease classification by leveraging ResNet50 and MobileNetV2 architectures","volume":"15","author":"Sharma","year":"2025","journal-title":"Sci Rep"},{"key":"B29","doi-asserted-by":"publisher","first-page":"123276","DOI":"10.1016\/j.eswa.2024.123276","article-title":"Resnet deep models and transfer learning technique for classification and quality detection of rice cultivars","volume":"247","author":"Razavi","year":"2024","journal-title":"Expert Syst Appl"},{"key":"B30","doi-asserted-by":"publisher","first-page":"228","DOI":"10.47852\/bonviewAIA3202853","article-title":"Comparison of deep learning techniques in detection of sickle cell disease","volume":"1","author":"Simon","year":"2023","journal-title":"Artif Intell Appl"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-025-59471-1","article-title":"Designer artificial environments for membrane protein synthesis","volume":"16","author":"Meyer","year":"2025","journal-title":"Nat Commun"},{"key":"B32","doi-asserted-by":"publisher","first-page":"8645","DOI":"10.1038\/s41598-024-58421-z","article-title":"The application of improved densenet algorithm in accurate image recognition","volume":"14","author":"Hou","year":"2024","journal-title":"Sci Rep"},{"key":"B33","doi-asserted-by":"publisher","first-page":"24455","DOI":"10.1007\/s00521-023-09038-2","article-title":"An adaptively weighted loss-enabled lightweight teacher\u2013student model for real-time railroad inspection on edge devices","volume":"35","author":"Guo","year":"2023","journal-title":"Neural Comput Appl"},{"key":"B34","doi-asserted-by":"publisher","first-page":"51","DOI":"10.1109\/MCE.2022.3181759","article-title":"Lightweight deep learning: an overview","volume":"13","author":"Wang","year":"2022","journal-title":"IEEE Consum Electron Mag"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge distillation: a survey","volume":"129","author":"Gou","year":"2021","journal-title":"Int J Comput Vis"},{"key":"B36","author":"Han","year":"2023"},{"key":"B37","doi-asserted-by":"publisher","first-page":"91","DOI":"10.54254\/2755-2721\/104\/20241142","article-title":"Diabetes prediction based on KNN, XGBoost, SVM and LR model","volume":"104","author":"Yang","year":"2024","journal-title":"Appl Comput Eng"},{"key":"B38","doi-asserted-by":"publisher","first-page":"1","DOI":"10.4108\/eetpht.9.3913","article-title":"An ensemble models for the prediction of sickle cell disease from erythrocytes smears","volume":"9","author":"Ayoade","year":"2023","journal-title":"EAI Endorsed Trans Pervasive Health Technol"},{"key":"B39","doi-asserted-by":"publisher","first-page":"39","DOI":"10.31449\/inf.v48i7.5726","article-title":"Comparative analysis of support vector machine, random forest and k-nearest neighbor classifiers for predicting remaining usage life of roller bearings","volume":"48","author":"Palaniappan","year":"2024","journal-title":"Informatica"},{"key":"B40","doi-asserted-by":"publisher","first-page":"yoaf020","DOI":"10.1093\/jscdis\/yoaf020","article-title":"Identification of hematological biomarkers and assessment of machine learning models for sickle cell anemia severity classification","volume":"2","author":"Ussher","year":"2025","journal-title":"J Sickle Cell Dis"},{"key":"B41","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1186\/s40537-024-00973-y","article-title":"Enhancing K-nearest neighbor algorithm: a comprehensive review and performance analysis of modifications","volume":"11","author":"Halder","year":"2024","journal-title":"J Big Data"},{"key":"B42","doi-asserted-by":"publisher","first-page":"69","DOI":"10.58496\/BJML\/2024\/007","article-title":"Random forest algorithm overview","volume":"2024","author":"Salman","year":"2024","journal-title":"Babyl J Mach Learn"},{"key":"B43","doi-asserted-by":"publisher","first-page":"414","DOI":"10.3390\/diagnostics16030414","article-title":"Toward AI-assisted sickle cell screening: a controlled comparison of CNN, transformer, and hybrid architectures using public blood-smear images","volume":"16","author":"Tasji","year":"2026","journal-title":"Diagnostics"},{"key":"B44","doi-asserted-by":"publisher","first-page":"23286","DOI":"10.1038\/s41598-026-53648-4","article-title":"Microscopic image enhancement for accurate sickle cell disease identification using intuitionistic fuzzy driven retinex","volume":"16","author":"Ramakrishnan","year":"2026","journal-title":"Sci Rep"},{"key":"B45","doi-asserted-by":"publisher","first-page":"5648","DOI":"10.3390\/s23125648","article-title":"An optimized deep learning model for predicting mild cognitive impairment using structural MRI","volume":"23","author":"Alyoubi","year":"2023","journal-title":"Sensors"},{"key":"B46","doi-asserted-by":"publisher","first-page":"13787","DOI":"10.1007\/s11042-024-19561-6","article-title":"Impact of optimizers functions on detection of melanoma using transfer learning architectures","volume":"84","author":"K\u0131l\u0131\u00e7arslan","year":"2025","journal-title":"Multimed Tools Appl"},{"key":"B47","doi-asserted-by":"publisher","first-page":"394","DOI":"10.36347\/sjet.2024.v12i12.008","article-title":"Application of AI algorithms for the prediction of the likelihood of sickle cell crises","volume":"12","author":"Okon","year":"2024","journal-title":"Sch J Eng Tech"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1938284\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:51:07Z","timestamp":1790661067000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1938284\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":47,"alternative-id":["10.3389\/fdgth.2026.1938284"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1938284","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]},"article-number":"1938284"}}