Advancements in medical imaging have significantly improved disease detection, segmentation, and classification, yet the computational demands of current deep-learning models often hinder their deployment in real-time clinical environments. This Methods Collection addresses these challenges by focusing on the development of efficient deep learning models, novel image enhancement techniques, and multimodal fusion approaches that can improve the accuracy, speed, and accessibility of medical imaging. By providing cutting-edge methodologies for lightweight architectures, image clarity enhancements, and the integration of multiple imaging modalities, this collection aims to enable the practical application of AI in resource-constrained settings such as mobile healthcare and edge computing devices. This will help bridge the gap between advanced AI research and real-world clinical applications, benefiting both researchers and clinicians in the ongoing effort to improve diagnostic processes and patient care.