Fortissimo Plus

The proposed project ImagingGenesis aims to develop a generative AI model focused on medical image analysis, specifically utilising latent diffusion models (LDM) to address challenges in organ-specific computed tomography (CT) image segmentation and reconstruction. This software will use the power of High-Performance Computing (HPC) resources to manage the computational intensity required for training advanced AI models. The primary objective is to demonstrate that LDMs can effectively mitigate the challenges posed by limited or incomplete datasets, a common issue in the medical imaging domain. By generating high-quality synthetic medical images, the diffusion model will complement existing datasets, improving the accuracy of future AI models made for specific organs. This project will lay the groundwork for enhancing CT imaging AI capabilities across various medical applications, proving that with the appropriate computational resources, the lack of sufficient data can be overcome. This endeavour will not only advance the field of medical image analysis but also set a precedent for future research and development of generative AI models in healthcare, enabling more precise diagnostics and treatment planning. The success of this project will be a significant step toward reducing dependency on large datasets, making AI-driven medical innovations more accessible and efficient.