GenerateCT: Text-Guided 3D Chest CT Generation

Welcome to the official page of GenerateCT. We are at the forefront of innovation in the field of text-conditional 3D medical image generation, with a special emphasis on chest CT volumes. This is where you will discover our collection of meticulously trained models for text-to-CT generation. Additionally, we provide a rich dataset of CT volumes that we've generated, each paired with corresponding text prompts. We are pleased to make these resources freely accessible to all researchers.Welcome to the official home of GenerateCT, where we are at the cutting edge of text-conditional 3D medical imaging, specifically tailored towards the generation of chest CT volumes. On our platform, you will have unrestricted access to our meticulously trained models dedicated to text-to-CT creation. Additionally, we provide a rich dataset of synthetically generated CT volumes, each paired with text prompts to aid your research. This invaluable resource is made freely available to all researchers in our continued commitment to advance the field of medical imaging.

Results

Our models achieve the following performances :

	Resolution	Dimension	Text-Guided	FID (↓)	FVD (↓)	CLIP (↑)
CT-ViT	128	3D	No	73.4	1817.4	N/A
Transformer	128	3D	Yes	104.3	1886.8	25.2
Diffusion	512	2D	Yes	14.9	409.8	27.6
GenerateCT	512	3D	Yes	55.8	1092.3	27.1

License

Our work, including the codes, trained models, and generated data, is released under a Creative Commons Attribution (CC-BY) license. This means that anyone is free to share (copy and redistribute the material in any medium or format) and adapt (remix, transform, and build upon the material) for any purpose, even commercially, as long as appropriate credit is given, a link to the license is provided, and any changes that were made are indicated. This aligns with our goal of facilitating progress in the field by providing a resource for researchers to build upon.