The field of medical image analysis and classification is rapidly evolving, with a focus on developing innovative methods to improve diagnostic accuracy and patient outcomes. Recent research has explored the adaptation of foundation models, such as the Segment Anything Model (SAM), for medical image classification, demonstrating their potential in capturing complex spatial and contextual details. Additionally, there has been a growing interest in weakly supervised pre-training frameworks, which have shown promising results in multi-instance learning for whole-slide pathology images. Continual learning approaches have also been proposed to address domain shifts and incremental learning in medical image analysis. Notably, some papers have introduced novel architectures and techniques, such as attention-based mechanisms and generative latent replay, to enhance the performance of deep learning models in medical image classification. The development of these methods has the potential to significantly impact clinical practice and improve patient care. Noteworthy papers include: Adapting a Segmentation Foundation Model for Medical Image Classification, which introduces a novel framework to adapt SAM for medical image classification. SimMIL: A Universal Weakly Supervised Pre-Training Framework for Multi-Instance Learning in Whole Slide Pathology Images, which proposes a weakly-supervised scheme for pre-training feature extractors. Attention-based Generative Latent Replay: A Continual Learning Approach for WSI Analysis, which employs Gaussian Mixture Models to synthesize WSI representations and patch count distributions. MIPHEI-ViT: Multiplex Immunofluorescence Prediction from H&E Images using ViT Foundation Models, which integrates state-of-the-art ViT foundation models to predict mIF signals from H&E images.