The field of medical image analysis is rapidly evolving, with a focus on developing innovative techniques for accurate diagnosis and treatment planning. Recent research has highlighted the potential of deep learning models, particularly convolutional neural networks (CNNs) and transformers, in improving the accuracy and efficiency of medical image analysis. These models have been applied to various tasks, including tumor classification, segmentation, and detection, with promising results. The integration of techniques such as explainable AI and diffusion models has further enhanced the robustness and interpretability of these models. Notably, the development of specialized frameworks and datasets has facilitated the advancement of research in this area. Some noteworthy papers include:
- A study that presented a comprehensive MRI tumor framework, achieving a test accuracy of 99.86% for brain tumor classification.
- A transformer-based framework for automated classification of breast lesions in dynamic contrast-enhanced MRI, achieving an AUC of 0.92 for lesion-level classification.
- A vision-guided diffusion model for brain tumor detection and segmentation, demonstrating consistent gains in Dice similarity and Hausdorff distance.