The Medical Imaging Analytics Platform provides a dedicated computational environment for developing, training, and validating deep learning models applied to medical imaging data. Built on NVIDIA RTX 6000 ADA and H100 NVL GPUs, MEDVISION is purpose-configured for the high memory and throughput demands of medical image processing, supporting standard imaging formats and pipelines used in clinical and research contexts. The platform supports a range of imaging modalities and analytical tasks including segmentation, classification, anomaly detection, and computer-aided diagnosis, using frameworks and architectures commonly applied in medical AI research. This testbed contributes to the CITADELS Framework by providing accessible, domain-focused infrastructure that supports the translation of medical imaging AI from research prototype to clinically relevant proof-of-concept.