Research Highlight

Events and News

Mohammad Farazi’s IPMI 2023 paper is Accepted, Congratulation! This paper employed the Volumetric Laplace Beltrami Operator over its conventionally used Graph Laplacian to build a convolutional neural network on tetrahedral meshes Inspired by ChebyNet. We adapted the cost function of localized minimum cut in the Graclus algorithm for downsampling the mesh based on LBO. Also, since we did not use any fixed template for the tetrahedral meshes as they have different sizes, we approximate the LBO on downsampled meshes using a simple piece-wise constant approximation scheme after each pooling. Finally, we used the ADNI dataset for both classifications and age prediction tasks to evaluate our model. We extended the conventionally used Grad-Cam algorithm to tetrahedral meshes for visualization to better interpret biomarkers.

Wenhui Zhu and Peijie Qiu’s IPMI 2023 paper is Accepted, Congratulation! Non-mydriatic retinal color fundus photography (CFP) is widely available due to the advantage of not requiring pupillary dilation, however, is prone to poor quality due to operators, systemic imperfections, or patient-related causes. Optimal retinal image quality is mandated for accurate medical diagnoses and automated analyses. Herein, we leveraged the Optimal Transport (OT) theory to propose an unpaired image-to-image translation scheme for mapping low-quality retinal CFPs to high-quality counterparts. Furthermore, to improve the flexibility, robustness, and applicability of our image enhancement pipeline in the clinical practice, we generalized a state-of-the-art model-based image re- construction method, regularization by denoising, by plugging in priors learned by our OT-guided image-to-image translation network.

Wenhui Zhu’s ISBI 2023 paper is Accepted, Congratulation! Real-world non-mydriatic retinal fundus photography is prone to artifacts, imperfections and low-quality when certain ocular or systemic co-morbidities exist. Artifacts may result in inaccuracy or ambiguity in clinical diagnoses. In this paper, we proposed a simple but effective end-to-end framework for enhancing poor-quality retinal fundus images. Leveraging the optimal transport theory, we proposed an unpaired image-to-image translation scheme for transporting low-quality images to their high-quality counterparts. We theoretically proved that a Generative Adversarial Net- works (GAN) model with a generator and discriminator is sufficient for this task.