AI-Powered Object Detection in Radiology: Current Models, Challenges, and Future Direction
Abstract
:1. Introduction
- The review of the existing AI models in the area of object detection in radiology, with particular reference to their application in diagnostic imaging modalities such as X-rays, MRI, CT scans, and US within the specific frameworks of CNNs, transformers, and hybrid architecture, is provided in the paper itself.
- Clinical applications of medical imaging object detection models are critically analyzed for potential benefits in terms of accuracy, robustness, and computational efficiency. Challenges concerning data scarcity, annotation quality, and the need for model generalization across population-based cohorts and different imaging protocols are discussed.
- The contribution reviews critical privacy and ethical issues in the application of deep learning object detection in radiology, and these must be the very basis for secure and compliant solutions required in health AI. It also provides a rationale behind the need to deal with biases in AI algorithms, making them treat all patients equally by healthcare systems.
- This review reveals new trends and future directions taken by AI research into radiology to assist collaboration among AI researchers, clinicians, and policymakers in bringing down the remaining barriers to the maximum benefit of AI techniques for improving diagnosis accuracy, patient outcomes, and modalities of medical imaging themselves.
2. Overview of Radiology Modalities
3. Importance of Imaging Preprocessing
4. Current AI Models for Object Detection
5. Data and Annotation Challenges
6. Real-World Implementation
7. Future Direction and Emerging Trends
7.1. Wearable Devices and Real-Time Monitoring
7.2. Regulator and Ethical Considerations
7.3. Advances in Deep Learning Algorithms
7.4. Explainable AI (XAI)
7.5. Cloud Computing and Big Data Analytics
7.6. Integrating AI-Powered Object Detection into Clinical Workflows
8. Conclusions
Author Contributions
Funding
Conflicts of Interest
References
- Elhanashi, A.; Lowe, D.; Saponara, S.; Moshfeghi, Y. Deep learning techniques to identify and classify COVID-19 abnormalities on chest X-ray images. In Proceedings of the Real-Time Image Processing and Deep Learning, Orlando, FL, USA, 3 April–13 June 2022; SPIE: Bellingham, WA, USA, 2022; Proceedings Volume 12102, p. 1210204. [Google Scholar] [CrossRef]
- McKinney, S.M.; Sieniek, M.; Godbole, V.; Godwin, J.; Antropova, N.; Ashrafian, H.; Back, T.; Chesus, M.; Corrado, G.S.; Darzi, A.; et al. International evaluation of an AI system for breast cancer screening. Nature 2020, 577, 89–94. [Google Scholar] [CrossRef] [PubMed]
- Topol, E.J. High-performance medicine: The convergence of human and artificial intelligence. Nat. Med. 2019, 25, 44–56. [Google Scholar] [CrossRef] [PubMed]
- Pesapane, F.; Codari, M.; Sardanelli, F. Artificial intelligence in medical imaging: Threat or opportunity? Radiologists again at the forefront of innovation in medicine. Eur. Radiol. Exp. 2018, 2, 35. [Google Scholar] [CrossRef] [PubMed]
- Hsu, C.-M.; Hsu, C.-C.; Hsu, Z.-M.; Shih, F.-Y.; Chang, M.-L.; Chen, T.-H. Colorectal polyp image detection and classification through grayscale images and deep learning. Sensors 2021, 21, 5995. [Google Scholar] [CrossRef]
- Wang, S.; Summers, R.M. Machine learning and radiology. Med. Image Anal. 2012, 16, 933–951. [Google Scholar] [CrossRef]
- Caruana, R.; Lou, Y.; Gehrke, J.; Koch, P.; Sturm, M.; Elhadad, N. Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sydney, Australia, 10–13 August 2015. [Google Scholar]
- Rajpurkar, P.; Irvin, J.; Zhu, K.; Yang, B.; Mehta, H.; Duan, T.; Ding, D.; Bagul, A.; Langlotz, C.; Shpanskaya, K.; et al. Chexnet: Radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv 2017, arXiv:1711.05225. [Google Scholar]
- Dong, X.; Xu, S.; Liu, Y.; Wang, A.; Saripan, M.I.; Li, L.; Zhang, X.; Lu, L. Multi-view secondary input collaborative deep learning for lung nodule 3D segmentation. Cancer Imaging 2020, 20, 53. [Google Scholar] [CrossRef]
- Isensee, F.; Petersen, J.; Klein, A.; Zimmerer, D.; Jaeger, P.F.; Kohl, S.; Wasserthal, J.; Koehler, G.; Norajitra, T.; Wirkert, S.; et al. nnu-net: Self-adapting framework for u-net-based medical image segmentation. arXiv 2018, arXiv:1809.10486. [Google Scholar]
- Ragab, M.G.; Abdulkadir, S.J.; Muneer, A.; Alqushaibi, A.; Sumiea, E.H.; Qureshi, R.; Al-Selwi, S.M.; Alhussian, H. A comprehensive systematic review of YOLO for medical object detection (2018 to 2023). IEEE Access 2024, 12, 57815–57836. [Google Scholar] [CrossRef]
- Zhao, Y.; Guo, Q.; Zhang, Y.; Zheng, J.; Yang, Y.; Du, X.; Feng, H.; Zhang, S. Application of deep learning for prediction of Alzheimer’s disease in PET/MR imaging. Bioengineering 2023, 10, 1120. [Google Scholar] [CrossRef]
- Onakpojeruo, E.P.; Mustapha, M.T.; Ozsahin, D.U.; Ozsahin, I. Enhanced MRI-based brain tumour classification with a novel Pix2pix generative adversarial network augmentation framework. Brain Commun. 2024, 6, fcae372. [Google Scholar] [CrossRef]
- Zhang, H.; Qie, Y. Applying deep learning to medical imaging: A review. Appl. Sci. 2023, 13, 10521. [Google Scholar] [CrossRef]
- Liu, W.; Zhang, B.; Liu, T.; Jiang, J.; Liu, Y. Artificial Intelligence in Pancreatic Image Analysis: A Review. Sensors 2024, 24, 4749. [Google Scholar] [CrossRef] [PubMed]
- Sani, Z.; Prasad, R.; Hashim, E.K.M. Breast Cancer Detection in Mammography using Faster Region Convolutional Neural Networks and Group Convolution. IETE J. Res. 2024, 70, 7379–7395. [Google Scholar] [CrossRef]
- Al-qaness, M.A.; Zhu, J.; AL-Alimi, D.; Dahou, A.; Alsamhi, S.H.; Abd Elaziz, M.; Ewees, A.A. Chest X-ray images for lung disease detection using deep learning techniques: A comprehensive survey. Arch. Comput. Methods Eng. 2024, 31, 3267–3301. [Google Scholar] [CrossRef]
- Walsh, J.; Othmani, A.; Jain, M.; Dev, S. Using U-Net network for efficient brain tumor segmentation in MRI images. Healthc. Anal. 2022, 2, 100098. [Google Scholar] [CrossRef]
- Nawaz, M.; Nazir, T.; Baili, J.; Khan, M.A.; Kim, Y.J.; Cha, J.H. CXray-EffDet: Chest disease detection and classification from X-ray images using the EfficientDet model. Diagnostics 2023, 13, 248. [Google Scholar] [CrossRef]
- Onakpojeruo, E.P.; Mustapha, M.T.; Ozsahin, D.U.; Ozsahin, I. A comparative analysis of the novel conditional deep convolutional neural network model, using conditional deep convolutional generative adversarial network-generated synthetic and augmented brain tumor datasets for image classification. Brain Sci. 2024, 14, 559. [Google Scholar] [CrossRef]
- Liu, J.; Zhang, Y.; Wang, K.; Yavuz, M.C.; Chen, X.; Yuan, Y.; Li, H.; Yang, Y.; Yuille, A.; Tang, Y.; et al. Universal and extensible language-vision models for organ segmentation and tumor detection from abdominal computed tomography. Med. Image Anal. 2024, 97, 103226. [Google Scholar] [CrossRef]
- Kanjanasurat, I.; Tenghongsakul, K.; Purahong, B.; Lasakul, A. CNN–RNN network integration for the diagnosis of COVID-19 using chest X-ray and CT images. Sensors 2023, 23, 1356. [Google Scholar] [CrossRef]
- Torghabeh, F.A.; Hosseini, S.A. Deep Learning-Based Brain Tumor Segmentation in MRI Images: A MobileNetV2-DeepLabV3+ Approach. Iran. J. Med. Phys. Majallah-I Fīzīk-I Pizishkī-i Īrān 2024, 21, 343–354. [Google Scholar]
- Sharma, P.; Balabantaray, B.K.; Bora, K.; Mallik, S.; Kasugai, K.; Zhao, Z. An ensemble-based deep convolutional neural network for computer-aided polyps identification from colonoscopy. Front. Genet. 2022, 13, 844391. [Google Scholar] [CrossRef] [PubMed]
- Balasubramanian, P.K.; Lai, W.C.; Seng, G.H.; Selvaraj, J. Apestnet with mask r-cnn for liver tumor segmentation and classification. Cancers 2023, 15, 330. [Google Scholar] [CrossRef] [PubMed]
- Aljabri, M.; AlAmir, M.; AlGhamdi, M.; Abdel-Mottaleb, M.; Collado-Mesa, F. Towards a better understanding of annotation tools for medical imaging: A survey. Multimed. Tools Appl. 2022, 81, 25877–25911. [Google Scholar] [CrossRef]
- Chan, H.P.; Samala, R.K.; Hadjiiski, L.M.; Zhou, C. Deep learning in medical image analysis. In Deep Learning in Medical Image Analysis: Challenges and Applications; Springer: Cham, Switzerland, 2020; pp. 3–21. [Google Scholar]
- Li, B.; Xu, Y.; Wang, Y.; Li, L.; Zhang, B. The student-teacher framework guided by self-training and consistency regularization for semi-supervised medical image segmentation. PLoS ONE 2024, 19, e0300039. [Google Scholar] [CrossRef]
- Pham, H.H.; Le, K.H.; Tran, T.V.; Nguyen, H.Q. Improving Object Detection in Medical Image Analysis through Multiple Expert Annotators: An Empirical Investigation. arXiv 2023, arXiv:2303.16507. [Google Scholar]
- Boafo, Y.G. An overview of computer—Aided medical image classification. Multimed. Tools Appl. 2024. [Google Scholar] [CrossRef]
- ElyElyan, E.; Vuttipittayamongkol, P.; Johnston, P.; Martin, K.; McPherson, K.; Moreno-García, C.F.; Jayne, C.; Sarker, M.K. Computer vision and machine learning for medical image analysis: Recent advances, challenges, and way forward. Artif. Intell. Surg. 2022, 2, 24–45. [Google Scholar]
- Moradi, M.; Keyvanpour, M.R. CAPTCHA for crowdsourced image annotation: Directions and efficiency analysis. Aslib J. Inf. Manag. 2022, 74, 522–548. [Google Scholar] [CrossRef]
- Saeed, S.U. Multi-Level Optimisation Using Deep Meta Learning for Medical Image Analysis. Ph.D. Thesis, UCL (University College London), London, UK, 2024. [Google Scholar]
- Heim, E.; Roß, T.; Seitel, A.; März, K.; Stieltjes, B.; Eisenmann, M.; Lebert, J.; Metzger, J.; Sommer, G.; Sauter, A.W. Large-scale medical image annotation with crowd-powered algorithms. J. Med. Imaging 2018, 5, 034002. [Google Scholar] [CrossRef]
- Juluru, K.; Shih, H.-H.; Murthy, K.N.K.; Elnajjar, P.; El-Rowmeim, A.; Roth, C.; Genereaux, B.; Fox, J.; Siegel, E.; Rubin, D.L. Integrating Al algorithms into the clinical workflow. Radiol. Artif. Intell. 2021, 3, e210013. [Google Scholar] [CrossRef] [PubMed]
- Yang, Y.; Zhang, H.; Gichoya, J.W.; Katabi, D.; Ghassemi, M. The limits of fair medical imaging AI in real-world generalization. Nat. Med. 2024, 30, 2838–2848. [Google Scholar] [PubMed]
- Schork, N.J. Artificial intelligence and personalized medicine. In Precision Medicine in Cancer Therapy; Springer: Cham, Switzerland, 2019; pp. 265–283. [Google Scholar]
- Lang, O.; Yaya-Stupp, D.; Traynis, I.; Cole-Lewis, H.; Bennett, C.R.; Lyles, C.R.; Lau, C.; Irani, M.; Semturs, C.; Webster, D.R.; et al. Using generative AI to investigate medical imagery models and datasets. EBioMedicine 2024, 102, 105075. [Google Scholar] [CrossRef] [PubMed]
- Koohi-Moghadam, M.; Bae, K.T. Generative AI in medical imaging: Applications, challenges, and ethics. J. Med. Syst. 2023, 47, 94. [Google Scholar] [CrossRef]
- Srivastava, S.; Bhatia, S.; Agrawal, A.P.; Jayswal, A.K.; Godara, J.; Dubey, G. Deep adaptive fusion with cross-modality feature transition and modality quaternion learning for medical image fusion. Evol. Syst. 2025, 16, 17. [Google Scholar]
- Tan, Z.; Yang, X.; Pan, T.; Liu, T.; Jiang, C.; Guo, X.; Wang, Q.; Nguyen, A.; Qi, A.; Huang, K.; et al. Personalize to generalize: Towards a universal medical multi-modality generalization through personalization. arXiv 2024, arXiv:2411.06106. [Google Scholar]
- Koçak, B.; Ponsiglione, A.; Stanzione, A.; Bluethgen, C.; Santinha, J.; Ugga, L.; Huisman, M.; Klontzas, M.E.; Cannella, R.; Cuocolo, R. Bias in artificial intelligence for medical imaging: Fundamentals, detection, avoidance, mitigation, challenges, ethics, and prospects. Diagn. Interv. Radiol. 2025, 31, 75. [Google Scholar] [CrossRef]
- Ganti, A.; Wilson, S.; Ma, Z.; Zhao, X.; Ma, R. Narrative detection and feature analysis in online health communities. In Proceedings of the 4th Workshop of Narrative Understanding (WNU2022), Seattle, WA, USA, 15 July 2022; pp. 57–65. [Google Scholar]
- Tang, J.; Shen, S.; Wang, Z.; Gong, Z.; Zhang, J.; Chen, X. When fairness meets bias: A debiased framework for fairness aware top-n recommendation. In Proceedings of the 17th ACM Conference on Recommender Systems, Singapore, 18–22 September 2023; pp. 200–210. [Google Scholar]
- Steimers, A.; Schneider, M. Sources of risk of AI systems. Int. J. Environ. Res. Public Health 2022, 19, 3641. [Google Scholar] [CrossRef]
- Tejani, A.S.; Ng, Y.S.; Xi, Y.; Rayan, J.C. Understanding and mitigating bias in imaging artificial intelligence. Radiographics 2024, 44, e230067. [Google Scholar]
- Chen, H. Design and implementation of secure enterprise network based on DMVPN. In Proceedings of the 2011 International Conference on Business Management and Electronic Information, Guangzhou, China, 13–15 May 2011. [Google Scholar]
- Gewida, M.H. Leveraging Machine Learning and Large Language Model to Mitigate Smart Home IoT Password Breaches. Ph.D. Thesis, Colorado Technical University, Colorado Springs, CO, USA, 2024. [Google Scholar]
- Singh, R.; Gill, S.S. Edge AI: A survey. Internet Things Cyber-Phys. Syst. 2023, 3, 71–92. [Google Scholar] [CrossRef]
- Adedoyin, M. Real-Time Healthcare Applications: Exploring the Synergy of Generative AI and Edge Computing/5G. In Modern Technologies in Healthcare; CRC Press: New York, NY, USA, 2025; pp. 212–240. [Google Scholar]
- Truong, H.-L.; Truong-Huu, T.; Cao, T.-D. Making distributed edge machine learning for resource-constrained communities and environments smarter: Contexts and challenges. J. Reliab. Intell. Environ. 2023, 9, 119–134. [Google Scholar]
- Pinto-Coelho, L. How artificial intelligence is shaping medical imaging technology: A survey of innovations and applications. Bioengineering 2023, 10, 1435. [Google Scholar] [CrossRef] [PubMed]
- Anand, A.; Krithivasan, S.; Roy, K. RoMIA: A framework for creating Robust Medical Imaging AI models for chest radiographs. Front. Radiol. 2024, 3, 1274273. [Google Scholar] [CrossRef] [PubMed]
- Alnaggar, O.A.M.F.; Jagadale, B.N.; Saif, M.A.N.; Ghaleb, O.A.; Ahmed, A.A.; Aqlan, H.A.A.; Al-Ariki, H.D.E. Efficient artificial intelligence approaches for medical image processing in healthcare: Comprehensive review, taxonomy, and analysis. Artif. Intell. Rev. 2024, 57, 221. [Google Scholar] [CrossRef]
- Apostolidis, K.D.; Papakostas, G.A. A survey on adversarial deep learning robustness in medical image analysis. Electronics 2021, 10, 2132. [Google Scholar] [CrossRef]
- Javed, H.; El-Sappagh, S.; Abuhmed, T. Robustness in deep learning models for medical diagnostics: Security and adversarial challenges towards robust AI applications. Artif. Intell. Rev. 2025, 58, 12. [Google Scholar] [CrossRef]
- Brasil, S.; Pascoal, C.; Francisco, R.; dos Reis Ferreira, V.; AVideira, P.; Valadão, G. Artificial intelligence (AI) in rare diseases: Is the future brighter? Genes 2019, 10, 978. [Google Scholar] [CrossRef]
- Mohammed, S.; Mohammed, N.; Sultana, W. A review of AI-powered diagnosis of rare diseases. Int. J. Curr. Sci. Res. Rev. 2024, 7. [Google Scholar] [CrossRef]
- Zhang, Z.; Citardi, D.; Wang, D.; Genc, Y.; Shan, J.; Fan, X. Patients’ perceptions of using artificial intelligence (AI)-based technology to comprehend radiology imaging data. Health Inform. J. 2021, 27, 14604582211011215. [Google Scholar] [CrossRef]
- Linguraru, M.G.; Bakas, S.; Aboian, M.; Chang, P.D.; Flanders, A.E.; Kalpathy-Cramer, J.; Kitamura, F.C.; Lungren, M.P.; Mongan, J.; Prevedello, L.M. Clinical, cultural, computational, and regulatory considerations to deploy AI in radiology: Perspectives of RSNA and MICCAI experts. Radiol. Artif. Intell. 2024, 6, e240225. [Google Scholar]
- Chen, M.; Zhang, B.; Cai, Z.; Seery, S.; Gonzalez, M.J.; Ali, N.M.; Ren, R.; Qiao, Y.; Xue, P.; Jiang, Y. Acceptance of clinical artificial intelligence among physicians and medical students: A systematic review with cross-sectional survey. Front. Med. 2022, 9, 990604. [Google Scholar] [CrossRef] [PubMed]
- Herington, J.; McCradden, M.D.; Creel, K.; Boellaard, R.; Jones, E.C.; Jha, A.K.; Rahmim, A.; Scott, P.J.H.; Sunderland, J.J.; Wahl, R.L. Ethical considerations for artificial intelligence in medical imaging: Deployment and governance. J. Nucl. Med. 2023, 64, 1509–1515. [Google Scholar] [CrossRef] [PubMed]
- Contaldo, M.T.; Pasceri, G.; Vignati, G.; Bracchi, L.; Triggiani, S.; Carrafiello, G. AI in radiology: Navigating medical responsibility. Diagnostics 2024, 14, 1506. [Google Scholar] [CrossRef] [PubMed]
- Smith, H. Clinical AI: Opacity, accountability, responsibility and liability. Ai Soc. 2021, 36, 535–545. [Google Scholar] [CrossRef]
- Latif, W.B.; Yasin, I.M.; Ali, M.J.; Islam, M.N.; Forid, M.S. Transforming Applied Medical Sciences: The Impact of AI, VR, and AR on Research, Education Technology, and Clinical Practices. J. Angiother. 2024, 8, 1–8. [Google Scholar]
- Venkatesan, M.; Mohan, H.; Ryan, J.R.; Schürch, C.M.; Nolan, G.P.; Frakes, D.H.; Coskun, A.F. Virtual and augmented reality for biomedical applications. Cell Rep. Med. 2021, 2, 100348. [Google Scholar] [CrossRef]
- US Food and Drug Administration. Augmented Reality and Virtual Reality in Medical Devices; US Food and Drug Administration: Silver Spring, MD, USA, 2023. [Google Scholar]
- Mbunge, E.; Muchemwa, B.; Jiyane, S.E.; Batani, J. Sensors and healthcare 5.0: Transformative shift in virtual care through emerging digital health technologies. Glob. Health J. 2021, 5, 169–177. [Google Scholar] [CrossRef]
- Bhamidipaty, V.; Bhamidipaty, D.L.; Guntoory, I.; Bhamidipaty, K.D.P.; Iyengar, K.P.; Botchu, B.; Botchu, R. Revolutionizing Healthcare: The Impact of AI-Powered Sensors. In Generative Artificial Intelligence for Biomedical and Smart Health Informatics; John Wiley & Sons: Hoboken, NJ, USA, 2025; pp. 355–373. [Google Scholar]
- Gautam, N.; Ghanta, S.N.; Mueller, J.; Mansour, M.; Chen, Z.; Puente, C.; Ha, Y.M.; Tarun, T.; Dhar, G.; Sivakumar, K.; et al. Artificial intelligence, wearables and remote monitoring for heart failure: Current and future applications. Diagnostics 2022, 12, 2964. [Google Scholar] [CrossRef]
- LaBoone, P.A.; Marques, O. Overview of the future impact of wearables and artificial intelligence in healthcare workflows and technology. Int. J. Inf. Manag. Data Insights 2024, 4, 100294. [Google Scholar] [CrossRef]
- Dankwa-Mullan, I. Health equity and ethical considerations in using artificial intelligence in public health and medicine. Prev. Chronic Dis. 2024, 21, E64. [Google Scholar] [CrossRef]
- Ooi, K. Using artificial intelligence in patient care—Some considerations for doctors and medical regulators. Asian Bioeth. Rev. 2024, 16, 483–499. [Google Scholar] [CrossRef] [PubMed]
- Pesapane, F.; Volonté, C.; Codari, M.; Sardanelli, F. Artificial intelligence as a medical device in radiology: Ethical and regulatory issues in Europe and the United States. Insights Into Imaging 2018, 9, 745–753. [Google Scholar] [CrossRef] [PubMed]
- Rigby, M.J. Ethical dimensions of using artificial intelligence in health care. AMA J. Ethics 2019, 21, 121–124. [Google Scholar]
- Razzak, M.I.; Naz, S.; Zaib, A. Deep learning for medical image processing: Overview, challenges and the future. In Classification in BioApps: Automation of Decision Making; Springer: Cham, Switzerland, 2017; pp. 323–350. [Google Scholar]
- Li, M.; Jiang, Y.; Zhang, Y.; Zhu, H. Medical image analysis using deep learning algorithms. Front. Public Health 2023, 11, 1273253. [Google Scholar] [CrossRef]
- Zhou, S.K.; Dou, Q.; Gao, Y.; Han, H.; Ma, J.; Sun, J.; Zhang, D.; Zhao, S.; Zheng, Y. Artificial Intelligence Algorithm Advances in Medical Imaging and Image Analysis. In Artificial Intelligence in Medical Imaging in China; Springer Nature: Singapore, 2024; pp. 83–110. [Google Scholar]
- Chen, D.; Mirebeau, J.M.; Shu, H.; Cohen, L.D. A region-based randers geodesic approach for image segmentation. Int. J. Comput. Vis. 2024, 132, 349–391. [Google Scholar] [CrossRef]
- Kumar, A.; Hora, H.; Rohilla, A.; Kumar, P.; Gautam, R. Explainable Artificial Intelligence (XAI) for Healthcare: Enhancing Transparency and Trust. In Proceedings of the International Conference on Cognitive Computing and Cyber Physical Systems, Delhi, India, 1–2 December 2023; Springer: Singapore, 2024. [Google Scholar]
- Rane, N.; Choudhary, S.; Rane, J. Explainable Artificial Intelligence (XAI) in Healthcare: Interpretable Models for Clinical Decision Support. 2023. Available online: https://ssrn.com/abstract=4637897 (accessed on 15 April 2025).
- Praveen, S.; Joshi, K. Explainable Artificial Intelligence in Health Care: How XAI Improves User Trust in High-Risk Decisions. In Explainable Edge AI: A Futuristic Computing Perspective; Springer International Publishing: Cham, Switzerland, 2022; pp. 89–99. [Google Scholar]
- Elhanashi, A.; Dini, P.; Saponara, S.; Zheng, Q. TeleStroke: Real-time stroke detection with federated learning and YOLOv8 on edge devices. J. Real-Time Image Proc. 2024, 21, 121. [Google Scholar] [CrossRef]
- Shakor, M.Y.; Khaleel, M.I. Recent Advances in Big Medical Image Data Analysis Through Deep Learning and Cloud Computing. Electronics 2024, 13, 4860. [Google Scholar] [CrossRef]
- Arepalli, B.R. The Transformative Impact of Cloud Computing and Big Data Analytics on Healthcare. Int. J. Sci. Res. 2024, 13, 1291–1295. [Google Scholar] [CrossRef]
- Chauhan, A.S.; Singh, R.; Priyadarshi, N.; Twala, B.; Suthar, S.; Swami, S. Unleashing the power of advanced technologies for revolutionary medical imaging: Pioneering the healthcare frontier with artificial intelligence. Discov. Artif. Intell. 2024, 4, 58. [Google Scholar] [CrossRef]
- Wu, K.; Wu, E.; Theodorou, B.; Liang, W.; Mack, C.; Glass, L.; Sun, J.; Zou, J. Characterizing the clinical adoption of medical AI devices through US insurance claims. AI 2024, 1, AIoa2300030. [Google Scholar] [CrossRef]
- Obuchowicz, R.; Strzelecki, M.; Piórkowski, A. Clinical applications of artificial intelligence in medical imaging and image processing—A review. Cancers 2024, 16, 1870. [Google Scholar] [CrossRef]
- Adler-Milstein, J.; Aggarwal, N.; Ahmed, M.; Castner, J.; Evans, B.J.; Gonzalez, A.A.; James, C.A.; Lin, S.; Mandl, K.D.; Matheny, M.E.; et al. Meeting the moment: Addressing barriers and facilitating clinical adoption of artificial intelligence in medical diagnosis. NAM Perspect. 2022, 2022, 10-31478. [Google Scholar] [CrossRef]
Modality | Description | Common Use Cases |
---|---|---|
X-ray Imaging | Uses electromagnetic radiation to obtain pictures of structures in the body, such as bones and lungs. | 1. Fracture diagnosis. 2. Diagnosis of chest abnormalities (e.g., pneumonia, tuberculosis). 3. Disease surveillance. |
Computed Tomography (CT) | Uses X-rays and a computer to produce cross-sectional views of the body’s anatomy. | 1. Tumor or cancer detection. 2. Vasculature imaging. 3. Trauma injuries imaging. |
Magnetic Resonance Imaging (MRI) | Uses magnetic fields and radio waves to produce detailed images of soft tissues. | 1. Disorders of the brain and spinal cord imaging. 2. Diagnosis of joint injuries. 3. Soft tissue tumor detection. |
Ultrasound | Utilizes high-pitched sound waves to provide real-time images of organ movement and blood flow. | 1. Fetal development observation. 2. Diagnosis of abdominal or pelvic conditions. 3. Examination of blood flow in arteries and veins. |
Positron Emission Tomography (PET) | Involves the injection of a radioactive tracer to visualize metabolic activity in tissues. | 1. Identification of cancer and monitoring treatment response. 2. Evaluation of brain disorders (e.g., Alzheimer’s). 3. Evaluation of heart function. |
X-ray Imaging | Uses electromagnetic radiation to obtain pictures of structures in the body, such as bones and lungs. | 1. Fracture diagnosis. 2. Diagnosis of chest abnormalities (e.g., pneumonia, tuberculosis). 3. Disease surveillance. |
Modality | AI Techniques Used | Key Contributions | References | Performance Metrics |
---|---|---|---|---|
X-ray | CNNs, Transfer Learning | Automated detection of lung diseases (e.g., COVID-19, tuberculosis) | [8] | Sensitivity: 90%, Specificity: 87% |
CT | 3D CNNs, Attention Mechanisms | Tumor segmentation and volume measurement | [9] | Coefficient: 0.85 |
MRI | GANs, U-Nets | Tumor detection, anomaly localization | [10] | Accuracy: 93%, Recall: 91% |
Ultrasound | YOLO, Region-based CNNs | Real-time detection of lesions, fetal abnormalities | [11] | Precision: 88%, F1 Score: 0.89 |
PET | Hybrid CNN-RNN Models | Early cancer detection, brain activity mapping | [12] | Sensitivity: 92%, Specificity: 89% |
Model | Application | Technique | Advantages | Limitations | Reference |
---|---|---|---|---|---|
YOLOv4 | Tumor Detection in CT scans | Deep Learning (CNN-based) | High accuracy, fast real-time detection | Limited interpretability, struggles with small tumors | [15] |
FasterR-CNN | Breast Cancer Detection in Mammography | Region Proposal Networks + CNN | High accuracy in bounding box prediction | Sensitive to noise, requires large datasets | [16] |
RetinaNet | Detection of Lung Nodules in Chest X-rays | Focal Loss with CNN | Handles class imbalance well | Performance drops with poor-quality images | [17] |
U-Net | Tumor Segmentation in Brain MRI | Convolutional Autoencoder (CNN) | High segmentation accuracy, pixel-level prediction | Overfitting on small datasets | [18] |
EfficientDet | Detection of Cardiac Abnormalities | EfficientNet-based Architecture | Efficient with limited resources | Reduced performance with complex image conditions | [19] |
Conditional (C-DCNN) | Brain Tumor Classification (MRI) | Conditional Deep CNN + GAN-generated synthetic data | Improved generalization with synthetic data, robust classification | Dependent on GAN quality, computational cost for training | [20] |
LLM-based Vision Transformers | Multi-organ Detection in CT/MRI | Transformer Architecture | Integrates multiple object detection tasks | High computational cost, needs large datasets | [21] |
Faster R-CNN + LSTM | Longitudinal Tracking of Tumor Progression in CT | CNN for Detection + LSTM for Sequence Modeling | Captures temporal changes in tumor progression | Complex, slow for real-time applications | [22] |
DeepLabV3+ | Detecting Brain Lesions in MRI scans | Atrous Convolution + CNN | High precision for lesion detection | Computationally expensive, requires fine-tuning | [23] |
Ensemble CNNs | Colon Polyp Detection in Colonoscopy images | Multiple CNN Architectures | Improved performance through ensemble learning | Computationally intensive, requires careful architecture selection | [24] |
Mask R-CNN | Liver Lesion Detection and Segmentation in CT images | Mask R-CNN + CNN | Accurate lesion segmentation and detection | Struggles with small lesions and low contrast images | [25] |
Challenge | Summary | Key Findings | Reference |
---|---|---|---|
Data Preparation | Extensive time required for data cleaning, formatting, and image selection. | Efficient preprocessing pipelines and automated tools can reduce preparation time and improve dataset quality. | [26] |
Annotator Expertise | Skilled annotators with expertise in annotation and healthcare are needed. | Training programs and collaboration with medical professionals can improve annotator expertise. | [27] |
Labeling Consistency | Consistency in labeling is crucial for reliable model training. | Standardized annotation protocols and regular quality checks can enhance labeling consistency. | [28] |
Limited Data Access | Limited access to diverse and representative medical images. | Data sharing agreements and federated learning approaches can improve data access while ensuring privacy. | [29] |
Data Bias | Bias in medical datasets can lead to inaccurate model predictions. | Ensuring diverse datasets and using techniques like data augmentation and synthetic data can address bias. | [30] |
Privacy Concerns | Patient privacy is a significant concern in medical image annotation. | Anonymization techniques and secure data handling are necessary to protect privacy. | [31] |
Annotation Tools | The choice of annotation tools impacts efficiency and accuracy. | Feature-rich, user-friendly tools can improve annotation efficiency and reduce errors. | [32] |
Quality Control | Regular quality checks are essential to maintain high annotation standards. | Multi-level review processes and automated quality control tools can enhance annotation reliability. | [33] |
Scalability | Scaling annotation efforts to large datasets is challenging. | Crowdsourcing platforms and scalable annotation frameworks can help manage large-scale projects. | [34] |
Integration with Clinical Workflows | Integrating annotation with clinical workflows enhances practical utility. | Collaboration with healthcare providers and tool integration into clinical systems can facilitate seamless data use. | [35] |
Category | Subcategory | Issue | Example | Potential Solutions |
---|---|---|---|---|
Computational Challenges | Model Generalization | Variability in image quality, resolution, and acquisition parameters across modalities [36] | Models trained on high-resolution MRI performed poorly on low-resolution MRI [37] | Multimodal architectures; domain adaptation techniques [38] |
Resource Constraints | Real-time processing exceeds hardware capabilities in resource-constrained environments [39] | AI-powered ultrasound systems face delays in low-resource settings [40] | Model optimization through quantization and pruning [41] | |
Ethical and Legal Challenges | Bias and Fairness | Imbalanced datasets lead to disparities in diagnostic accuracy across populations [42] | AI for breast cancer detection underperformed for African-descent patient [43] | Diverse datasets; fairness-aware training strategies [44] |
Accountability and Liability | Undefined responsibility in AI-related errors [45] | AI stroke detection misclassification delayed treatment [46] | Comprehensive AI usage guidelines; standardized error analysis protocols [47] | |
Integration Challenges | Interoperability | Incompatibility of AI tools with health systems like PACS and EHR [48] | CT anomaly detection system failed to integrate with legacy PACS [49] | Standards adherence (DICOM, HL7 FHIR); collaboration with vendors |
Clinician Acceptance | Lack of trust due to AI’s black-box nature [50] | Skepticism about lung nodule detection AI due to limited explainability [51] | Focus on explainability, user-friendly interfaces [52] | |
Performance Challenges | Robustness to Variations | Inconsistent imaging conditions like motion artifacts and positioning impact performance [53] | High false positives in X-rays with motion artifacts [54] | Augmented datasets; preprocessing for normalization [55] |
Handling Rare Cases | Poor detection of rare conditions due to underrepresentation in training [56] | AI failed to identify rare glioblastoma subtype [57] | Synthetic data generation; transfer learning [58] | |
Economic Challenges | High Costs | High initial investments deter smaller facilities [59] | AI imaging system upgrade cost over USD 1 million for a tertiary hospital [60] | Public–private partnerships; subsidized solutions [61] |
Workforce Training | Training lack of AI familiarity hampers effective usage [62] | Radiologists needed six months to learn AI diagnostic tools [63] | Comprehensive training programs; AI in medical education [64] |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Elhanashi, A.; Saponara, S.; Zheng, Q.; Almutairi, N.; Singh, Y.; Kuanar, S.; Ali, F.; Unal, O.; Faghani, S. AI-Powered Object Detection in Radiology: Current Models, Challenges, and Future Direction. J. Imaging 2025, 11, 141. https://doi.org/10.3390/jimaging11050141
Elhanashi A, Saponara S, Zheng Q, Almutairi N, Singh Y, Kuanar S, Ali F, Unal O, Faghani S. AI-Powered Object Detection in Radiology: Current Models, Challenges, and Future Direction. Journal of Imaging. 2025; 11(5):141. https://doi.org/10.3390/jimaging11050141
Chicago/Turabian StyleElhanashi, Abdussalam, Sergio Saponara, Qinghe Zheng, Nawal Almutairi, Yashbir Singh, Shiba Kuanar, Farzana Ali, Orhan Unal, and Shahriar Faghani. 2025. "AI-Powered Object Detection in Radiology: Current Models, Challenges, and Future Direction" Journal of Imaging 11, no. 5: 141. https://doi.org/10.3390/jimaging11050141
APA StyleElhanashi, A., Saponara, S., Zheng, Q., Almutairi, N., Singh, Y., Kuanar, S., Ali, F., Unal, O., & Faghani, S. (2025). AI-Powered Object Detection in Radiology: Current Models, Challenges, and Future Direction. Journal of Imaging, 11(5), 141. https://doi.org/10.3390/jimaging11050141