Medical Imaging (Segmentation and Reconstruction)

Our work introduces a novel, low-cost, and mobile 3D imaging method utilizing gradient illumination, specifically designed for effective dermatological evaluation, potentially revolutionizing early diagnosis and monitoring of skin diseases using everyday devices.

Our study introduces a pioneering two-step, domain-specific approach to enhance sparse-view CT reconstruction, employing a Sinogram Inpainting Network and a Postprocessing Refining Network coupled with a Discriminator Perceptual loss, significantly improving reconstruction quality and efficiency.

We propose an iterative reconstruction approach to reduce artifacts, such as beam hardening and dispersion effects, in Talbot-Lau X-ray grating interferometry computed tomography by incorporating a polychromatic forward projection model that includes prior knowledge about the physical setup.

Hui Lin, Santiago López-Tapia, Florian Schiffers, Y. Wu, S. Gunasekaran, J. Hwang, D. Bishara, E. Kholmovski, M. Elbaz, RS. Passman, Daniel Kim, and Aggelos K. Katsaggelos Heliyon 2024 Fig. 1.