Kushtia: Two students of Islamic University (IU) in Kushtia have secured 11th place globally in the preliminary round of the GAVE-2 (Generalized Analysis of Vessels in Eye, Edition-2) Challenge at the 29th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI). They have also been invited to attend the MICCAI 2026 conference in person in Strasbourg, France.
According to Bangladesh Sangbad Sangstha, a total of 247 teams participated in the GAVE-2 challenge. Following the preliminary evaluation, the top 15 teams were selected for the final round, with Team PixelPulse securing 11th position. The finalists, Jubaer Rahman and Faisal Ahmed, are currently pursuing master's degrees from the Information and Communication Technology department under the 2022-23 session. Their team, PixelPulse, has been selected among the top 15 teams for the final round of the prestigious international challenge.
The challenge featured three tasks. The first task involved using different machine learning models to identify and segment retinal arteries and veins separately from Colour Fundus (CF) images. The second task required participants to analyse Colour Fundus and Fundus Fluorescein Angiography (FFA) images together to separately identify retinal arteries and veins. In the third task, participants had to analyse the results obtained from Task 2 to quantify retinal vascular biomarkers, including the width of major arteries and veins, their relationship, and the proportion of retinal area occupied by the vessels.
Team leader Jubaer Rahman expressed his excitement, noting, 'MICCAI was my dream conference. I am extremely happy to have been selected for on-site participation. I am proud to represent the ICT Department and Islamic University on such a prestigious global research platform.' He added that researchers from leading universities worldwide will attend the conference, and he hopes to attend the main conference in France if there are no visa issues.
Team member Faisal Ahmed highlighted the global competition, stating that many universities and research institutions from around the world participated in the MICCAI 2026 GAVE-2 Challenge. 'After competing with them, Team PixelPulse from Bangladesh was the only team from the country selected for the final round. We are working on retinal blood vessel analysis using artificial intelligence and machine learning models, which could have significant applications in medical image analysis in the future.' He concluded by expressing pride in achieving this recognition with limited resources.