Abstract

Chest X-ray diagnosis is delayed because of the increasing workload faced by radiologists and the intrinsic tendency of deep learning models to produce false positive detections. This paper proposes a lightweight multi-modal deep learning framework to address these challenges. Potential disease regions are identified using a YOLOv5 object detector, and a binary classifier based on an EfficientNet architecture with transfer learning is employed. The main role of the classifier is to remove the detector’s false positive predictions. On the VinDr-CXR benchmark (18k chest X-rays, 14 abnormality types, expert bounding boxes), the proposed system obtains a mean average precision (mAP) of 0.246. This results in a relative improvement of 69% over the baseline YOLOv5 detector (mAP 0.145) and is significantly better than the performance of the standalone YOLOv5 (mAP 0.145) and Faster R-CNN (mAP 0.142). Using a two-stage cascade of a YOLOv5 detector and an EfficientNet-based classifier with transfer learning, the accuracy of disease localization in chest X-rays is significantly improved. This approach improves the reliability of computer-aided diagnostic workflows by reducing false positives and does not require large-scale pretraining or complex transformer architectures.

References

• Sultan Ahmad, Stephen Neal Joshua Eali, Thirupathi Rao Nakka, Naceur Chihaoui, Tahani Alsubait, and Humaira Khanam. A scalable and reliable deep learning framework for enhanced brain tumor detection and diagnosis using ai-based medical imaging. Frontiers in Medicine, 13, 2026.
• Saad Albawi, Tareq Abed Mohammed, and Saad Al- Zawi. Understanding of a convolutional neural network. In 2017 International Conference on Engineering and Technology (ICET), pages 1–6. Ieee, 2017.
• Pamela G. Anderson, Hannah Tarder-Stoll, Mehmet Alpaslan, Nora Keathley, David L. Levin, Srivas Venkatesh, Elliot Bartel, Serge Sicular, Scott Howell, Robert V. Lindsey, and Rebecca M. Jones. Deep learning improves physician accuracy in the comprehensive detection of abnormalities on chest X-rays. Scientific Reports, 14:25151, 2024.
• Talha Anwar and Seemab Zakir. Effect of image augmentation on ecg image classification using deep learning. In 2021 International Conference on Artificial Intelligence (ICAI), pages 182–186, 2021.
• Yi-Ching Cheng, Yi-Chieh Hung, Guan-Hua Huang, Tai-Been Chen, Nan-Han Lu, Kuo-Ying Liu, and Kuo-Hsuan Lin. Deep learning-based object detection strategies for disease detection and localization in chest X-ray images. Diagnostics, 14(23):2636, 2024.
• Mohamed Elgendi, Muhammad Umer Nasir, Qunfeng Tang, David Smith, John-Paul Grenier, Catherine Batte, Bradley Spieler, William Donald Leslie, Carlo Menon, Richard Ribbon Fletcher, et al. The effectiveness of image augmentation in deep learning networks for detecting covid-19: A geometric transformation perspective. Frontiers in Medicine, 8, 2021.
• Weijie Fan, Yi Yang, Jing Qi, Qichuan Zhang, Cuiwei Liao, Li Wen, Shuang Wang, Guangxian Wang, Yu Xia, Qihua Wu, Xiaotao Fan, Xingcai Chen, Mi He, Jingjing Xiao, Liu Yang, Yun Liu, Jia Chen, Bing Wang, Lei Zhang, Liuqing Yang, Hui Gan, Shushu Zhang, Guofang Liu, Xiaodong Ge, Yuanqing Cai, Gang Zhao, Xi Zhang, Mingxun Xie, Huilin Xu, Yi Zhang, Jiao Chen, Jun Li, Shuang Han, Ke Mu, Shilin Xiao, Tingwei Xiong, Yongjian Nian, and Dong Zhang. A deep-learning-based framework for identifying and localizing multiple abnormalities and assessing cardiomegaly in chest X-ray. Nature Communications, 15:1347, 2024.
• Ruihua Guo, Kalpdrum Passi, and Chakresh Kumar Jain. Tuberculosis diagnostics and localization in chest x-rays via deep learning models. Frontiers in Artificial Intelligence, 3:583427, 2020.
• Shengnan Hao, Xinlei Li,Wei Peng, Zhu Fan, Zhanlin Ji, and Ivan Ganchev. Yolo-cxr: A novel detection network for locating multiple small lesions in chest x-ray images. IEEE Access, 12:156003–156019, 2024.
• Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition, 2015.
• Jeremy Irvin, Pranav Rajpurkar, Michael Ko, Yifan Yu, Silviana Ciurea-Ilcus, Chris Chute, Henrik Marklund, Behzad Haghgoo, Robyn Ball, Katie Shpanskaya, et al. Chexpert: A large chest radiograph dataset with uncertainty labels and expert comparison. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 590–597, 2019.
• Peiyuan Jiang, Daji Ergu, Fangyao Liu, Ying Cai, and Bo Ma. A review of yolo algorithm developments. Procedia Computer Science, 199:1066–1073, 2022.
• Saddam Hussain Khan, Anabia Sohail, Muhammad Mohsin Zafar, and Asifullah Khan. Coronavirus disease analysis using chest x-ray images and a novel deep convolutional neural network. Photodiagnosis and Photodynamic Therapy, 35:102473, 2021.
• Cherry Khosla and Baljit Singh Saini. Enhancing performance of deep learning models with different data augmentation techniques: A survey. In 2020 International Conference on Intelligent Engineering and Management (ICIEM), pages 79–85, 2020.
• H.E. Kim, H.H. Kim, B.K. Han, E.S. Ko, and K.J. Nam. Verification networks for false-positive reduction in medical object detection. Medical Image Analysis, 75:102272, 2022.
• Victoria Kramer, Irina Papazova, Andreas Thoma, Miriam Kunz, Peter Falkai, Thomas Schneider- Axmann, Anke Hierundar, Elias Wagner, and Alkomiet Hasan. Subjective burden and perspectives of german healthcare workers during the covid-19 pandemic. European archives of psychiatry and clinical neuroscience, 271:271–281, 2021.
• Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In European conference on computer vision, pages 740–755. Springer, 2014.
• Daniel A Moses. Deep learning applied to automatic disease detection using chest x-rays. Journal of Medical Imaging and Radiation Oncology, 65(5):498–517, 2021.
• Mamillapally Nagaraju, Priyanka Chawla, and Neeraj Kumar. Performance improvement of deep learning models using image augmentation techniques. Multimedia Tools and Applications, 81(7):9177–9200, 2022.
• Dillip Ranjan Nayak, Neelamadhab Padhy, Pradeep Kumar Mallick, Mikhail Zymbler, and Sachin Kumar. Brain tumor classification using dense efficient-net. Axioms, 11(1):34, 2022.
• Ha Q. Nguyen, Khanh Lam, Linh T. Le, Hieu H. Pham, Dat Q. Tran, Dung B. Nguyen, Dung D. Le, Chi M. Pham, Hang T. T. Tong, Diep H. Dinh, Cuong D. Do, Luu T. Doan, Cuong N. Nguyen, Binh T. Nguyen, Que V. Nguyen, Au D. Hoang, Hien N. Phan, Anh T. Nguyen, Phuong H. Ho, Dat T. Ngo, Nghia T. Nguyen, Nhan T. Nguyen, Minh Dao, and Van Vu. Vindr-cxr: An open dataset of chest xrays with radiologist’s annotations, 2021.
• Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, highperformance deep learning library. Advances in neural information processing systems, 32:8026–8037, 2019.
• B.N. Patel, L. Rosenberg, G. Willcox, D. Baltaxe, M. Lyons, et al. Confidence calibration for object detection in chest x-rays. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 621–631. Springer, 2023.
• Stephen M Pizer, E Philip Amburn, John D Austin, Robert Cromartie, Ari Geselowitz, Trey Greer, Bart ter Haar Romeny, John B Zimmerman, and Karel Zuiderveld. Adaptive histogram equalization and its variations. Computer vision, graphics, and image processing, 39(3):355–368, 1987.
• Petar Przulj. ChestGPT: Integrating large language models and vision transformers for disease detection and localization in chest X-rays. In Proceedings of the Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), 2025. arXiv:2507.03739.
• Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. You only look once: Unified, real-time object detection, 2016.
• Mustafain Rehman, Qiao Pan, Dehua Chen, and Arslan Manzoor. Classification of chest x-ray images to detect pneumonia using cnn and transfer learning. In 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pages 3498–3504, 2021.
• Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks, 2016.
• Hamid Rezatofighi, Nathan Tsoi, JunYoung Gwak, Amir Sadeghian, Ian Reid, and Silvio Savarese. Generalized intersection over union: A metric and a loss for bounding box regression. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 658–666, 2019.
• Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.
• Joanne Spetz. There are not nearly enough nurses to handle the surge of coronavirus patients: Here’s how to close the gap quickly. Health Affairs Forefront, 2020.
• John M. Statheros, Hairong Wang, and Richard Klein. CLARiTy: A vision transformer for multilabel classification and weakly-supervised localization of chest X-ray pathologies. In arXiv preprint arXiv:2512.16700, 2024. arXiv:2512.16700.
• Mingxing Tan and Quoc Le. Efficientnet: Rethinking model scaling for convolutional neural networks. In International Conference on Machine Learning, pages 6105–6114. PMLR, 2019.
• Y. Tang, R. Gao, T. Han, C. Chen, S. Zhang, and S.K. Zhou. Attention-guided multiple instance learning for anomaly localization in chest radiographs. IEEE Transactions on Medical Imaging, 43(2):678– 690, 2024.
• Janice Hopkins Tanne, Erika Hayasaki, Mark Zastrow, Priyanka Pulla, Paul Smith, and Aser Garcia Rada. Covid-19: how doctors and healthcare systems are tackling coronavirus worldwide. Bmj, 368, 2020.
• Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, Mohammadhadi Bagheri, and Ronald M Summers. Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 2097–2106, 2017.
• Z. Wang, Y. Liu, K. Chen, W. Zhang, Y. Zhou, A.L. Yuille, et al. Cxr-foundation: A general-purpose foundation model for chest radiography interpretation and localization. Nature Medicine, 31(2):345– 356, 2025.
• Zefan Yang, Xuanang Xu, Jiajin Zhang, Ge Wang, Mannudeep K. Kalra, and Pingkun Yan. Chest Xray foundation model with global and local representations integration. arXiv preprint arXiv:2502.05142, 2025. arXiv:2502.05142.
• Tahmina Zebin and Shahadate Rezvy. Covid-19 detection and disease progression visualization: Deep learning on chest x-rays for classification and coarse localization. Applied Intelligence, 51(2):1010–1021, 2021.
• R. Zhang, Z. Xu, S.H. Park, K.H. Moon, and J.G. Lee. Semi-supervised object detection for chest xray abnormality localization. Computerized Medical Imaging and Graphics, 105:102201, 2023.
• Weihang Zhang, Yuma Kinoshita, and Hitoshi Kiya. Image-enhancement-based data augmentation for improving deep learning in image classification problem. In 2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan), pages 1–2, 2020.
• Xinyu Zhang, Lijun Liu, Xiaobing Yang, Li Liu, and Wei Peng. Nsec-yolo: Real-time lesion detection on chest x-ray with adaptive noise suppression and global perception aggregation. Journal of Radiation Research and Applied Sciences, 2025.
• Fatema Tuz Zohora and K. C. Santosh. Circular foreign object detection in chest x-ray images. In K.C. Santosh, Mallikarjun Hangarge, Vitoantonio Bevilacqua, and Atul Negi, editors, Recent Trends in Image Processing and Pattern Recognition, pages 391–401, Singapore, 2017. Springer Singapore.