Paper Infomation
Artificial Intelligence in Medical Image‑Aided Diagnosis: An Overview
Full Text(PDF, 413KB)
Author: Beibei Hu, Liang Zhou, Qiong Ni
Abstract: This paper provides a comprehensive review of artificial intelligence (AI) applications in medical imaging diagnosis and assesses its clinical translation challenges. Through systematic review and meta-analysis, we evaluated AI diagnostic performance across imaging modalities including ophthalmology, respiratory medicine, and breast imaging. Results show that AI models consistently achieve high sensitivity and specificity, with AUC values above 0.9. However, real-world implementation remains limited due to methodological weaknesses, bias issues, and lack of external validation. The study emphasizes the need for a shift from accuracy-focused development to trustworthy, fair, and sustainable AI systems. We propose a novel framework for integrating AI into clinical practice through transparent reporting, human-AI collaboration, and rigorous validation, offering practical guidance for future AI deployment in healthcare.
Keywords: Medical Imaging, Clinical Translation, Trustworthy AI, Diagnosis
References:
[1] Obuchowicz, R., Lasek, J., Wodziński, M. et al. “Artificial Intelligence-Empowered Radiology-Current Status and Critical Review.” Diagnostics (Basel) 15 (2025): 282. doi:10.3390/diagnostics15030282
[2] Andreas Rowald, Salif Komi, Robin Demesmaeker. et al. (2022). Activity-dependent spinal cord neuromodulation rapidly restores trunk and leg motor functions after complete paralysis. Nature Medicine. https://doi.org/10.1038/s41591-021-01423-5
[3] Al-Obeidat, Feras, Wael Hafez, Asrar Rashid. et al. “Artificial Intelligence for the Detection of Acute Myeloid Leukemia from Microscopic Blood Images: A Systematic Review and Meta-Analysis.” Frontiers in Big Data 7 (2024): 1402926. doi:10.3389/fdata.2024.1402926.
[4] Xipeng Pan, Jijun Cheng, Feihu Hou. et al. (2023). “SMILE: Cost-sensitive Multi-task Learning for Nuclear Segmentation and Classification with Imbalanced Annotations.” Medical Image Analysis 88 (2023): 102867. https://doi.org/10.1016/j.media.2023.102867.
[5] Yunfang Yu, Wei Ren, Luhui Mao. et al.(2025). MRI-based multimodal AI model enables prediction of recurrence risk and adjuvant therapy in breast cancer. https://doi.org/10.1016/j.phrs.2025.107765
[6] Han, X., Yan, Q., Gao, N. et al. (2025) “Biocompatible Lyotropic Nanocarriers for Improved Delivery of Ascorbyl Tetraisopalmitate in Skincare.” Langmuir 41
[7] Chen, F., Wang, L., Hong, J. et al. “Unmasking Bias in Artificial Intelligence: A Systematic Review of Bias Detection and Mitigation Strategies in Electronic Health Record-Based Models.” Journal of the American Medical Informatics Association 31 (2024): 1172-83. doi:10.1093/jamia/ocae060.
[8] Zhang, H., Jian, M., Shao, M. et al.“Cross Spatio-Temporal Pathology-based Lung Nodule Dataset.” Scientific Data 11 (2024): 1007. Accessed. doi:10.1038/s41597-024-03851-7
[9] Shahid, G. S., Ahmad, J., Warraich, C. A. et al. “LIU-NET: Lightweight Inception U-Net for Efficient Brain Tumor Segmentation from Multimodal 3D MRI Images.” PeerJ Computer Science 11 (2025): e2787. doi:10.7717/peerj-cs.2787.
[10] Wang, Z., Lan, Y., Xu, Z.. et al. “Scientific Data Sharing and Reuse: Comparison of Mortality Predictive Models of Sepsis Patients Based on Machine Learning.” Chinese Medical Sciences Journal 37, no. 3 (2022): 201–10. https://doi.org/10.24920/004102.
[11] Chang, Y.-W., Ryu, J. K., An, J. K. et al. “Artificial Intelligence for Breast Cancer Screening in Mammography (AI-STREAM): Preliminary Analysis of a Prospective Multicenter Cohort Study.” Nature Communications (2025). https://doi.org/10.1038/s41467-025-57469-3.
[12] Hall, Jeff. “FDA Clears AI Software for Assessing Intracerebral Hemorrhage on Non-Contrast CT.” Diagnostic Imaging, February 12, 2024. https://www.diagnosticimaging.com/view/fda-ai-software-intracerebral-hemorrhage-non-contrast-ct.
[13] Katharina Wenderott, Jim Krups, Matthias Weigl. et al. (2024). Effects of artificial intelligence implementation on efficiency in medical imaging—a systematic literature review and meta-analysis. npj Digital Medicine. https://doi.org/10.1038/s41746-024-01248-9
[14] Besler, M. S., Koç, U. “Systematic Review of Artificial Intelligence Competitions in Radiology: A Focus on Design, Evaluation, and Trends.” Diagn Interv Radiol (2025). https://doi.org/10.4274/dir.2025.243152.
[15] Gupte, T., Nitave, T., Gobburu, J. et al. “Regulatory Landscape of Accelerated Approval Pathways for Medical Devices in the United States and the European Union.” Frontiers in Medical Technology 7 (2025): 1586070. https://doi.org/10.3389/fmedt.2025.1586070.
[16] Wenderott, K., Krups, J., Weigl, M. et al. “Facilitators and Barriers to Implementing AI in Routine Medical Imaging: Systematic Review and Qualitative Analysis.” Journal of Medical Internet Research 27 (2025): e63649. https://doi.org/10.2196/63649.
[17] Soleimantabar, H., Mahdavi, A., Qorbani, M. et al. “Artificial Intelligence in Radiology: Perceptions, Adoption Barriers, and Trust Among Iranian Radiologists in a Global Context.” InfoScience Trends 2 (2025): 1–12. https://doi.org/10.61186/ist.202502.03.01.