Welcome to Intellimaging Tech

We have proposed a novel score-matching formula to derive a new score function through deep learning on an image dataset. By integrating our new score function into the image reconstruction process, we have developed a new iterative reconstruction method within the MAP estimation framework to enhance image quality. We have evaluated the performance of our image reconstruction method on both public medical CT datasets and clinical raw datasets. Our reconstruction method consistently produced higher quality images in terms of PSNR and SSIM metrics across diverse datasets. Notably, on Siemens and GE clinical CT raw datasets, our proposed approach achieved superior denoising and deblurring effects over the competing methods, illustrating remarkable generalizability and stability. Our proposed score matching formula holds potential in image denoising, deblurring, and generation.

Our focus is on medical imaging, with an emphasis on X-ray tomographic imaging, photoacoustic imaging, and image reconstruction and analysis. We are developing theories, methods, software, and hardware systems for clinical applications. We are excited to collaborate with partners and funding agencies on innovative and impactful projects, especially National Institutes of Health.