HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments.

HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments | medRxiv Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments View ORCID Profile Simeon Mayala , Deogratias Mzurikwao , Emmanuel Suluba doi: https://doi.org/10.64898/2026.03.16.26348570 Simeon Mayala 1 Department of Biomedical Engineering, School of Biomedical Sciences, Muhimbili University of Health and Allied Sciences , Dar es salaam, Tanzania 3 Emerging Technology For Healthcare (ETH) Lab, East African Centre of Excellence for Cardiovascular Sciences (EACoECVS), Muhimbili University of Health and Allied Sciences , Dar es salaam, Tanzania Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Simeon Mayala For correspondence: mayalasimeon{at}gmail.com Deogratias Mzurikwao 1 Department of Biomedical Engineering, School of Biomedical Sciences, Muhimbili University of Health and Allied Sciences , Dar es salaam, Tanzania 3 Emerging Technology For Healthcare (ETH) Lab, East African Centre of Excellence for Cardiovascular Sciences (EACoECVS), Muhimbili University of Health and Allied Sciences , Dar es salaam, Tanzania Find this author on Google Scholar Find this author on PubMed Search for this author on this site Emmanuel Suluba 2 Department of Anatomy, School of Biomedical Sciences, Muhimbili University of Health and Allied Sciences , Dar es salaam, Tanzania Find this author on Google Scholar Find this author on PubMed Search for this author on this site Abstract Full Text Info/History Metrics Data/Code Preview PDF Abstract Deep learning model classification on large datasets is often limited in countries with restricted computational resources. While transfer learning can offset these limitations, standard architectures often maintain a high memory footprint. This study introduces HybridNet-XR, a memory-efficient and computationally lightweight hybrid convolutional neural network (CNN) designed to bridge the domain gap in medical radiography using autonomous self-supervised learning protocols. The HybridNet-XR architecture integrates depthwise separable convolutions for parameter reduction, residual connections for gradient stability, and aggressive early downsampling to minimize the video RAM (VRAM) footprint. We evaluated several training paradigms, including teacher-free self-supervised learning (SSL-SimCLR), teacher-led knowledge distillation (KD), and domain-gap (DG) adaptation. Each variant was pre-trained on ImageNet-1k subsets and fine-tuned on the ChestX6 multi-class dataset. Model interpretability was validated through gradient-weighted class activation mapping (Grad-CAM). The performance frontier analysis identified the HybridNet-XR-150-PW (Pre-warmed) as the optimal configuration, achieving a 93.38% average accuracy and 99% AUC while utilizing only 814.80 MB of VRAM. Regarding class-wise accuracy, this variant significantly outperformed standard MobileNetV2 and teacher-led models in critical diagnostic categories, notably Covid-19 (97.98%) and Emphysema (96.80%). Grad-CAM visualizations confirmed that the teacher-free pre-warming phase allows the model to develop sharper, anatomically grounded focus on pathological landmarks compared to distilled models. Specialized pre-warming schedules offer a viable, computationally autonomous alternative to knowledge distillation for medical imaging. By eliminating the requirement for high-performance teacher models, HybridNet-XR provides a robust and trustworthy diagnostic foundation suitable for clinical deployment in resource-constrained environments. Author summary Traditional deep learning models for medical imaging are often too large for the low-power computers available in many global health settings. We developed a new model to bridge this computational gap. We designed HybridNet-XR, a highly efficient AI architecture, and trained it using a “teacher-free” method that doesn’t require a massive supercomputer. We found a specific version ( H - XR 150 - PW ) that provides high accuracy while using very little memory. Our results show that high-performance diagnostic AI can be deployed on standard, low-cost hardware. Furthermore, using visual heatmaps (Grad-CAM), we proved that the AI correctly identifies medical landmarks like lung opacities, ensuring it is safe and reliable for real-world clinical use. Competing Interest Statement The authors have declared no competing interest. Funding Statement The author(s) received no specific funding for this work. Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: Not applicable I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data Availability All the datasets used in this study are freely available online https://www.kaggle.com/datasets/ammarkhanjadoon/chest-xray-14-dataset https://www.kaggle.com/datasets/mohamedasak/chest-x-ray-6-classes-dataset https://www.kaggle.com/datasets?search=Imagenet Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Back to top Previous Next Posted March 19, 2026. Download PDF Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments Simeon Mayala , Deogratias Mzurikwao , Emmanuel Suluba medRxiv 2026.03.16.26348570; doi: https://doi.org/10.64898/2026.03.16.26348570 Share This Article: Copy Citation Tools HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained Environments Simeon Mayala , Deogratias Mzurikwao , Emmanuel Suluba medRxiv 2026.03.16.26348570; doi: https://doi.org/10.64898/2026.03.16.26348570 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Health Informatics Subject Areas All Articles Addiction Medicine (638) Allergy and Immunology (917) Anesthesia (348) Cardiovascular Medicine (4976) Dentistry and Oral Medicine (493) Dermatology (426) Emergency Medicine (674) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1683) Epidemiology (16216) Forensic Medicine (36) Gastroenterology (1238) Genetic and Genomic Medicine (7257) Geriatric Medicine (751) Health Economics (1093) Health Informatics (5263) Health Policy (1463) Health Systems and Quality Improvement (1826) Hematology (599) HIV/AIDS (1376) Infectious Diseases (except HIV/AIDS) (16469) Intensive Care and Critical Care Medicine (1199) Medical Education (697) Medical Ethics (156) Nephrology (748) Neurology (7554) Nursing (377) Nutrition (1120) Obstetrics and Gynecology (1293) Occupational and Environmental Health (1038) Oncology (3733) Ophthalmology (1087) Orthopedics (407) Otolaryngology (464) Pain Medicine (499) Palliative Medicine (143) Pathology (720) Pediatrics (1854) Pharmacology and Therapeutics (752) Primary Care Research (785) Psychiatry and Clinical Psychology (6105) Public and Global Health (9961) Radiology and Imaging (2516) Rehabilitation Medicine and Physical Therapy (1506) Respiratory Medicine (1264) Rheumatology (657) Sexual and Reproductive Health (799) Sports Medicine (595) Surgery (799) Toxicology (109) Transplantation (318) Urology (307)

Related documents

• A Surgical Foundation Model Reveals Task-Dependent Label Efficiency#1115777
• Mitigating Spurious Invariance in Contrastive Learning: A Probabilistic Self-Supervised Learning Framework for Medical Image Analysis.#406453
• An Effective Ensemble Multimodal Learning System for Medical Diagnosis#183753
• Artificial Intelligence and Early Detection of Pancreatic Cancer#686473
• Vision-language model-based semantic-guided imaging biomarker for lung nodule malignancy prediction.#301616