1 code implementation • 16 Mar 2023 • Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Salah A. Baker, Stewart Lee Zuckerbrod
To address these issues, we propose SwinVFTR, a new transformer-based architecture designed for precise fluid segmentation in 3D volumetric OCT images.
1 code implementation • 24 Jun 2022 • Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Stewart Lee Zuckerbrod, Salah A. Baker
Ophthalmic images may contain identical-looking pathologies that can cause failure in automated techniques to distinguish different retinal degenerative diseases.
2 code implementations • 14 Apr 2021 • Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Stewart Lee Zuckerbrod, Salah A. Baker
The only non-invasive method for capturing retinal vasculature is optical coherence tomography-angiography (OCTA).
2 code implementations • 3 Jan 2021 • Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Stewart Lee Zuckerbrod, Kenton M. Sanders, Salah A. Baker
In order to avoid the loss of fidelity suffered by traditional GAN-based segmentation systems, we introduce a novel weighted feature matching loss.
Ranked #2 on Retinal Vessel Segmentation on STARE
Generative Adversarial Network Retinal Vessel Segmentation +2
1 code implementation • 11 May 2020 • Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Stewart Lee Zuckerbrod, Salah A. Baker, Kenton M. Sanders
Angiography requires insertion of a dye that may cause severe adverse effects and can even be fatal.
Ranked #6 on Fundus to Angiography Generation on Fundus Fluorescein Angiogram Photographs & Colour Fundus Images of Diabetic Patients
Fundus to Angiography Generation Generative Adversarial Network