Reproduce “ViBMR: A Large-Scale Multi-Sequence Brain MRI Dataset and Comprehensive Evaluation for Vietnamese Radiology Report Generation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dai Lam Le, Minh Khoi Nguyen, Son Dinh Nguyen, Amir Reza Jafari, Viet Ha, Nguyen Son, Reza Farahbakhsh, Noel Crespi and 1 more
ViBMR provides routine Vietnamese brain MRI findings paired with 3,649 patients and 11,875 series across four sequences. The authors test segmentation-pretrained encoders that fuse available sequence features for language-model report generation. On cross-dataset evaluation, BLEU-4 falls from 13.68 on MR-RATE to 1.03 on ViBMR, indicating domain shift without identifying its cause; fusion results vary by encoder and metric.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.