Skip to content

IReL_IIT(BHU) at MEDIQA-CORE-Task-2 2026: Hybrid Multimodal Learning for Radiology Report Discrepancy Analysis

2026 · Conference and Labs of the Evaluation Forum · pp. 4016-4028 · 1 citation · 18 references
Computer Science

TL;DR

IReL_IIT(BHU)’s submission to the MEDIQA-CORE Task 2 at CLEF 2026, focusing on automated agreement classification between radiology reports and suggested edits is presented, proposing a hybrid multimodal framework integrating transformer-based biomedical language modeling, multimodal feature fusion, hierarchical reasoning, synthetic data augmentation, and rule-based clinical inference.

View source

Similar papers

Review 2026

Overview of the MEDIQA-CORE 2026 Task 2 on Radiology Report Discrepancy Assessment

Results demonstrate that carefully designed prompting strategies can compete with substantially more complex image-grounded architectures, while multimodal methods continue to provide complementary clinical context for discrepancy understanding.

Asma Ben Abacha, Zhao-Yi Sun, Wen-wai Yim et al. · 3 citations
2026

NeuroVietCH at MEDIQA-CORE-Task-1 2026: LLM-Guided Multimodal Distillation for Brain Tumor Subtype Classification

Results suggest that LLM-guided distillation can effectively integrate clinical text reasoning with multimodal visual embeddings for robust brain tumor subtype classification and suggest that LLM-guided distillation can effectively integrate clinical text reasoning with multimodal visual embeddings for robust brain tum...

Sy-Duy Dang, D. Tran, Ngoc-Vuong Chu et al. · 1 citation
Open access 2026

MADS-VQA: MacroMamba and Dual-Sequence Feature Fusion for Medical Visual Question Answering

Medical Visual Question Answering (Med-VQA) aims to generate answers from medical images and natural language questions, supporting clinical decision-making and intelligent healthcare. However, existing methods often rely on homogeneous semantic modeling, limiting their ability to capture global anatomical structures a...

Xu Zhao, Ya-Ting Yue, Meng Han et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.