News

Towards Large-Scale 3D Medical Image Pre-training with Geometric Context Priors

We introduce the 3D Medical Vision Foundation Model VoCo. The proposed method aims to leverage geometric context priors for self-supervision. This paper also introduces a new benchmark and investigates the scaling laws of medical foundation models.

Last updated on 2024/10/16

[Academic Achievement] UG Students at HKUST SMART Lab Won Tse Cheuk Ng Tai Scholarship 2024

HKUST CSE UG student Mr. Runsheng LIU was selected as 2023-2024's Tse Cheuk Ng Tai Scholarship recipient with his UROP study titled 'Deep learning for medical image analysis'. Congratulations!

Last updated on 2024/09/27

[IEEE-TNNLS] Reference-based OCT Angiogram Super-resolution with Learnable Texture Generation

A new study on Optical Coherence Tomography Angiography (OCTA) image super-resolution from the HKUST Smart Lab has been accepted by the IEEE Transactions on Neural Networks and Learning Systems.

Last updated on 2024/09/19

[IEEE-TIP] Rethinking Self-training for Semi-supervised Landmark Detection: A Selection-free Approach

A new study on computational pathology from HKUST Smart Lab on semi-supervised landmark detection has been accepted by IEEE Transactions on Image Processing.

Last updated on 2024/09/16

[IEEE-TMI] Shapley Values-enabled Progressive Pseudo Bag Augmentation for Whole Slide Image Classification

A new study on computational pathology from HKUST Smart Lab on progressive pseudo bag augmentation based on Shapley values has been accepted by IEEE Transactions on Medical Imaging.

Last updated on 2024/09/14