{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T03:00:18Z","timestamp":1777950018712,"version":"3.51.4"},"reference-count":51,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T00:00:00Z","timestamp":1777852800000},"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. Artif. Intell."],"abstract":"<jats:p>Accurate detection of pediatric fractures in radiographs remains challenging due to subtle visual cues and the high prevalence of false-positive detections produced by automated systems. To address this limitation, we propose a lightweight region-of-interest (Region of Interest) adjudication framework that operates as a second-stage verification module to refine detector-generated candidates. The proposed framework integrates iterative hard-negative mining with confidence-aware score fusion to suppress anatomically confounding regions such as growth plates and overlapping structures. Unlike end-to-end detection approaches, the method is designed to function as a modular post-detection refinement stage, enabling improved decision reliability without modifying the underlying detector architecture. Each candidate Region of Interest is evaluated using a compact adjudication network conditioned on detector confidence, and final predictions are obtained through a calibrated fusion strategy. The framework is evaluated on the publicly available GRAZPEDWRI-DX pediatric radiograph dataset using patient-level disjoint training, validation, and held-out test splits to ensure unbiased performance estimation. Experimental results demonstrate that the proposed approach reduces false-positive detections while maintaining high sensitivity. At the selected operating point, the method achieves an F1-score of 0.88 and mAP@0.5 of 0.887, outperforming the detector-only baseline under identical evaluation conditions. In addition, gradient-based activation mapping (Grad-CAM) is employed to provide Region of Interest-level visual explanations, supporting interpretability of adjudication decisions. The proposed framework maintains low computational overhead, making it suitable for integration into real-world clinical workflows as a decision-support component.<\/jats:p>","DOI":"10.3389\/frai.2026.1807189","type":"journal-article","created":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T05:34:02Z","timestamp":1777872842000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A lightweight region of interest-level adjudication framework with hard-negative mining and confidence-aware fusion for pediatric fracture detection"],"prefix":"10.3389","volume":"9","author":[{"given":"C. V.","family":"Aravinda","sequence":"first","affiliation":[{"name":"Department of Computing and Electronics Engineering, Middle East College (Affiliated With Coventry University)","place":["Muscat, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Noushath","family":"Shaffi","sequence":"additional","affiliation":[{"name":"Department of Computer Science, College of Science, Sultan Qaboos University","place":["Muscat, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vimbi","family":"Viswan","sequence":"additional","affiliation":[{"name":"College of Computing and Information Science, University of Technology and Applied Sciences","place":["Sohar, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Adham","family":"Al-Rahbi","sequence":"additional","affiliation":[{"name":"Anatomical Pathology Program, Oman Medical Specialty Board","place":["Muscat, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Choy","family":"Ker Woon","sequence":"additional","affiliation":[{"name":"Department of Anatomy, Faculty of Medicine, Universiti Teknologi MARA","place":["Sungai Buloh, Malaysia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yassine","family":"Bouchareb","sequence":"additional","affiliation":[{"name":"Department of Radiology and Molecular Imaging, College of Medicine and Health Sciences, Sultan Qaboos University","place":["Muscat, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Srinivasa Rao","family":"Sirasanagandla","sequence":"additional","affiliation":[{"name":"Department of Human and Clinical Anatomy, College of Medicine and Health Sciences, Sultan Qaboos University","place":["Muscat, Oman"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,4]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2212","DOI":"10.3390\/diagnostics15172212","article-title":"Fracfusionnet: A multi-level feature fusion convolutional network for bone fracture detection in radiographic images","volume":"15","author":"Abd El-Ghany","year":"2025","journal-title":"Diagnostics"},{"key":"B2","doi-asserted-by":"publisher","first-page":"521","DOI":"10.1038\/s41597-023-02432-4","article-title":"FracAtlas: a dataset for fracture classification, localization and segmentation of musculoskeletal radiographs","volume":"10","author":"Abedeen","year":"2023","journal-title":"Sci. Data"},{"key":"B3","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1186\/s12880-023-01084-5","article-title":"Medical image diagnosis based on adaptive hybrid quantum cnn","volume":"23","author":"Ajlouni","year":"2023","journal-title":"BMC Med. Imaging"},{"key":"B4","doi-asserted-by":"publisher","first-page":"e15734056309890","DOI":"10.2174\/0115734056309890240912054616","article-title":"An integrated approach using yolov8 and resnet, seresnet &vision transformer (vit) algorithms based on roi fracture prediction in x-ray images of the elbow. Curr. Med","volume":"20","author":"Alam","year":"2024","journal-title":"Imag"},{"key":"B5","first-page":"1","article-title":"Deep learning for fracture detection in wrist radiographs. Sci","volume":"10","author":"Amr","year":"2020","journal-title":"Rep"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02873","article-title":"\u201cDycon: dynamic uncertainty-aware consistency and contrastive learning for semi-supervised medical image segmentation,\u201d","author":"Assefa","year":"2025","journal-title":"Proceedings of the Computer Vision and Pattern Recognition Conference"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1109\/BIBM62325.2024.10821983","article-title":"\u201cA comparative analysis of xai techniques for medical imaging: challenges and opportunities,\u201d","author":"Barra","year":"2024","journal-title":"2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)"},{"key":"B8","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1001\/jama.2009.50","article-title":"Mortality risk associated with low-trauma osteoporotic fracture and subsequent fracture in men and women","volume":"301","author":"Bliuc","year":"2009","journal-title":"JAMA"},{"key":"B9","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1109\/TMI.2020.3022034","article-title":"Lesion-harvester: Iteratively mining unlabeled lesions and hard-negative examples at scale. IEEE Trans","volume":"40","author":"Cai","year":"2020","journal-title":"Med. Imaging"},{"key":"B10","doi-asserted-by":"publisher","first-page":"5574","DOI":"10.1364\/BOE.529505","article-title":"Mflunet: multi-scale fusion lightweight unet for medical image segmentation. Biomed","volume":"15","author":"Cao","year":"2024","journal-title":"Opt. Express"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN54540.2023.10191869","article-title":"\u201cSegqnas: quantum-inspired neural architecture search applied to medical image semantic segmentation,\u201d","author":"Carlos","year":"2023","journal-title":"2023 International joint conference on neural Networks (IJCNN)"},{"key":"B12","first-page":"1","article-title":"\u201cCompressedmediq: hybrid quantum machine learning pipeline for high-dimensional neuroimaging data,\u201d","volume-title":"2025 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","author":"Chen","year":"2025"},{"key":"B13","article-title":"\u201cAutomatic detection of wrist fractures from radiographs using deep learning,\u201d","volume-title":"Medical Imaging 2021: Computer-Aided Diagnosis","author":"Cicero","year":"2021"},{"key":"B14","first-page":"6","article-title":"Assessing artificial intelligence and radiologist performance in musculoskeletal fracture detection: Multi-reader, multi-case study. Artif. Intell","volume":"26","author":"Dand\u00e1r","year":"2025","journal-title":"Med. Healthcare"},{"key":"B15","article-title":"Lightweight g-yolov11: advancing efficient fracture detection in pediatric wrist x-rays","author":"Ferdi","year":"2024","journal-title":"arXiv preprint arXiv:2501.00647"},{"key":"B16","first-page":"e210042","article-title":"Deep learning assessment of musculoskeletal radiographs: a systematic review and meta-analysis","volume":"3","author":"Gichoya","year":"2021","journal-title":"Radiology"},{"key":"B17","doi-asserted-by":"publisher","first-page":"11425","DOI":"10.1007\/s00521-024-09672-4","article-title":"An enhanced vision transformer with scale-aware and spatial-aware attention for thighbone fracture detection. Neur. Comput","volume":"36","author":"Guan","year":"2024","journal-title":"Applic"},{"key":"B18","first-page":"1321","article-title":"\u201cOn calibration of modern neural networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Guo","year":"2017"},{"key":"B19","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1186\/1471-227X-6-4","article-title":"Errors in fracture diagnoses in the emergency department-characteristics of patients and diurnal variation. BMC Emerg","volume":"6","author":"Hallas","year":"2006","journal-title":"Med"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e70260","DOI":"10.1002\/mco2.70260","article-title":"Artificial intelligence in orthopedic surgery: current applications, challenges, and future directions","volume":"6","author":"Han","year":"2025","journal-title":"MedComm"},{"key":"B21","doi-asserted-by":"publisher","first-page":"887","DOI":"10.1002\/jbmr.4814","article-title":"Deep-learning-based detection of vertebral fracture and osteoporosis using lateral spine x-ray radiography. J. Bone Miner","volume":"38","author":"Hong","year":"2020","journal-title":"Res"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72384-1_2","article-title":"\u201cA clinical-oriented multi-level contrastive learning method for disease diagnosis in low-quality medical images,\u201d","author":"Hou","year":"2024","journal-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140","article-title":"\u201cSearching for mobilenetv3,\u201d","author":"Howard","year":"2019","journal-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision"},{"key":"B24","doi-asserted-by":"publisher","first-page":"3207","DOI":"10.3390\/diagnostics13203207","article-title":"Discrepancies between radiology specialists and residents in fracture detection from musculoskeletal radiographs","volume":"13","author":"Huhtanen","year":"2023","journal-title":"Diagnostics"},{"key":"B25","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/INCET64471.2025.11140007","article-title":"\u201cEnhancing skin cancer diagnosis with xai: Integrating grad-cam, shap, and lime for model interpretability,\u201d","volume-title":"2025 6th International Conference for Emerging Technology (INCET)","author":"Jadhav","year":"2025"},{"key":"B26","doi-asserted-by":"publisher","first-page":"20077","DOI":"10.1038\/s41598-023-47460-7","article-title":"Fracture detection in pediatric wrist trauma x-ray images using yolov8 algorithm. Sci","volume":"13","author":"Ju","year":"2023","journal-title":"Rep"},{"key":"B27","doi-asserted-by":"publisher","first-page":"e13248","DOI":"10.1049\/ell2.13248","article-title":"Yolov9 for fracture detection in pediatric wrist trauma x-ray images. Electr","volume":"60","author":"Ju","year":"2024","journal-title":"Lett"},{"key":"B28","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1038\/s41526-024-00364-w","article-title":"Sans-CNN: an automated machine learning technique for spaceflight associated neuro-ocular syndrome with astronaut imaging data","volume":"10","author":"Kamran","year":"2024","journal-title":"NPJ Microgr"},{"key":"B29","article-title":"Brain tumor diagnosis using quantum convolutional neural networks","author":"Khan","year":"2024","journal-title":"arXiv preprint arXiv:2401.15804"},{"key":"B30","doi-asserted-by":"publisher","first-page":"983","DOI":"10.1080\/1206212X.2024.2404082","article-title":"A review of explainable ai in medical imaging: implications and applications. Int. J. Comput","volume":"46","author":"Kinger","year":"2024","journal-title":"Applic"},{"key":"B31","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/ICTAS64866.2025.11155355","article-title":"\u201cAutomated breast cancer classification from medical imaging using mobilenet,\u201d","volume-title":"2025 Conference on Information Communications Technology and Society (ICTAS)","author":"Kubakisa","year":"2025"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1879","DOI":"10.3390\/diagnostics14171879","article-title":"Artificial intelligence-based applications for bone fracture detection using medical images: a systematic review","volume":"14","author":"Kutbi","year":"2024","journal-title":"Diagnostics"},{"key":"B33","doi-asserted-by":"publisher","first-page":"4325","DOI":"10.1002\/mp.16244","article-title":"Region-related focal loss for 3D brain tumor MRI segmentation. Med","volume":"50","author":"Li","year":"2023","journal-title":"Phys"},{"key":"B34","article-title":"Mombs: Mixed-order minibatch sampling enhances model training from diverse-quality images","author":"Li","year":"2025","journal-title":"arXiv preprint arXiv:2505.18741"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324","article-title":"\u201cFocal loss for dense object detection,\u201d","author":"Lin","year":"2017","journal-title":"Proceedings of the IEEE International Conference on Computer Vision"},{"key":"B36","doi-asserted-by":"publisher","first-page":"11591","DOI":"10.1073\/pnas.1806905115","article-title":"Deep neural network improves fracture detection by clinicians. Proc. Nat. Acad","volume":"115","author":"Lindsey","year":"2018","journal-title":"Sci"},{"key":"B37","doi-asserted-by":"publisher","first-page":"e0318542","DOI":"10.1371\/journal.pone.0318542","article-title":"Advancing malware imagery classification with explainable deep learning: A state-of-the-art approach using shap, lime and grad-cam","volume":"20","author":"Nazim","year":"2025","journal-title":"PLoS ONE"},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.1145\/3368555.3384468","article-title":"\u201cHidden stratification causes clinically meaningful failures in machine learning for medical imaging,\u201d","author":"Oakden-Rayner","year":"2020","journal-title":"Proceedings of the ACM Conference on Health, Inference, and Learning"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-025-11296-6","article-title":"A comprehensive review of ai methods in upper extremity\/limb bone fracture detection. Artif. Intell","author":"Pour","year":"2025","journal-title":"Rev"},{"key":"B40","author":"Rauber","year":"2021","journal-title":"Grazpedwri-dx: A pediatric musculoskeletal radiograph dataset for computer-aided fracture detection and localization"},{"key":"B41","doi-asserted-by":"publisher","first-page":"5305","DOI":"10.3390\/s25175305","article-title":"Multi-scale attention networks with feature refinement for medical item classification in intelligent healthcare systems","volume":"25","author":"Riaz","year":"2025","journal-title":"Sensors"},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74","article-title":"\u201cGrad-CAM: visual explanations from deep networks,\u201d","author":"Selvaraju","year":"2017","journal-title":"Proceedings of the IEEE International Conference on Computer Vision"},{"key":"B43","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1051\/sicotj\/2023018","article-title":"Artificial intelligence for fracture diagnosis in orthopedic x-rays: current developments and future potential","volume":"9","author":"Sharma","year":"2023","journal-title":"SICOT-J"},{"key":"B44","doi-asserted-by":"publisher","first-page":"879","DOI":"10.3390\/bioengineering12080879","article-title":"Beyond post hoc explanations: a comprehensive framework for accountable AI in medical imaging through transparency, interpretability, and explainability","volume":"12","author":"Singh","year":"2025","journal-title":"Bioengineering"},{"key":"B45","doi-asserted-by":"publisher","first-page":"1295","DOI":"10.1016\/j.surg.2020.07.038","article-title":"Artificial intelligence in trauma systems","volume":"169","author":"Stonko","year":"2021","journal-title":"Surgery"},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1109\/ACOIT62457.2024.10939330","article-title":"\u201cQICCNN: quantum-inspired complex convolutional neural network based brain tumor detection from MRI images,\u201d","author":"Subhashini","year":"2024","journal-title":"2024 Asian Conference on Intelligent Technologies (ACOIT)"},{"key":"B47","doi-asserted-by":"crossref","first-page":"1179","DOI":"10.1109\/AUTOCOM64127.2025.10957315","article-title":"\u201cFractureai revolutionizing bone fracture detection with deep learning cnns for precision medical imaging,\u201d","volume-title":"2025 International Conference on Automation and Computation (AUTOCOM)","author":"Sudha","year":"2025"},{"key":"B48","doi-asserted-by":"publisher","first-page":"1135","DOI":"10.3390\/life15071135","article-title":"Automatic fracture detection convolutional neural network with multiple attention blocks using multi-region x-ray data","volume":"15","author":"Sumon","year":"2025","journal-title":"Life"},{"key":"B49","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/APCI65531.2025.11137192","article-title":"\u201cEnhanced bone fracture detection through deep learning-based multi-scale feature fusion,\u201d","volume-title":"2025 International Conference on Advancements in Power, Communication and Intelligent Systems (APCI)","author":"Sushma","year":"2025"},{"key":"B50","doi-asserted-by":"publisher","first-page":"4793","DOI":"10.1109\/TNNLS.2020.3027314","article-title":"A survey on explainable artificial intelligence (XAI): toward medical xai. IEEE Trans. Neural Netw. Learn","volume":"32","author":"Tjoa","year":"2020","journal-title":"Syst"},{"key":"B51","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1109\/TAI.2022.3153754","article-title":"Ensemble image explainable AI (XAI) algorithm for severe community-acquired pneumonia and COVID-19 respiratory infections. IEEE Trans. Artif","volume":"4","author":"Zou","year":"2022","journal-title":"Intell"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1807189\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T05:34:05Z","timestamp":1777872845000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1807189\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,4]]},"references-count":51,"alternative-id":["10.3389\/frai.2026.1807189"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1807189","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,4]]},"article-number":"1807189"}}