Artificial intelligence (AI) has emerged as a transformative force in radiology, enabling automated MRI and CT interpretation with unprecedented accuracy and efficiency. This comprehensive review expands on the technical principles, deep learning models, clinical applications, workflow integration, and existing limitations associated with AI in radiological imaging. By examining multimodal datasets, segmentation architectures, evaluation strategies, and regulatory perspectives, the article provides a detailed foundation for understanding how AI systems are currently utilized in neuroradiology, thoracic imaging, stroke diagnostics, and cardiovascular imaging. The review emphasizes the need for robust validation, bias mitigation, explainable AI, and integration with PACS/RIS platforms for safe and effective clinical adoption.

Artificial intelligence, deep learning, MRI, CT, radiology, automated diagnosis, segmentation, neural networks

Erickson BJ, Korfiatis P, Akkus Z. Artificial Intelligence in Medical Imaging. Radiology. 2024;310(2):423-438.

Nair A et al. Enhancing Radiologist Productivity with AI in MRI. Diagnostics. 2025.

Gillies R, Kinahan P, Hricak H. Radiomics and Imaging Biomarkers. Radiology. 2023.

Park SH, Han K. Clinical AI Evaluation Guidelines. Radiology. 2023.

Esteva A et al. Deep Learning in Clinical Diagnosis. Nature Medicine. 2024.

Liu X, Jiang T. AI for Brain Tumor MRI Classification. Medical Physics. 2025.

Wong KCL, Wood B. AI in Stroke Imaging. Stroke. 2024.