SearcharxivSearch

arXiv subjects

Quang-Huy Tran

Publications and source records attributed to Quang-Huy Tran.

2 recordsLinked to original sources

Integrating Unimodal and Vision-Language Representations in Latent Space for Multi-Label Chest X-Ray Classification

Multi-label chest X-ray classification has attracted considerable attention in recent years, with the effective use of visual representations and clinical semantic knowledge playing an important role. This study proposes a framework that combines unimodal representations from RAD-DINO with vision--language representations from BioViL-T for the classification of 14 labels in the MIMIC-CXR-JPG dataset. The RAD-DINO and BioViL-T embeddings and their combined representation are refined separately in latent space before being normalized and fused across the three branches. In addition to improving classification performance, the study aims to clarify the role of each embedding source and the degree to which they complement one another. Experiments show that RAD-DINO outperforms BioViL-T when used independently, whereas early fusion further improves the results, indicating that the two embedding sources contain complementary information. The best-performing model achieves a mean AUROC of 0.840 and an mAP of 0.467. Ablation analysis shows that hybrid fusion provides consistent and statistically significant improvements over early fusion when each embedding source is refined in latent space, suggesting that fusion effectiveness depends on the quality of the representation supplied by each branch. However, the study has only been evaluated internally on MIMIC-CXR-JPG; its generalizability to data from other healthcare institutions therefore remains to be validated. The source code is available at: https://anonymous.4open.science/r/mimic-report-C210/.

cs.CV

Upstream mobility Finite Volumes for the Richards equation in heterogenous domains

This paper is concerned with the Richards equation in a heterogeneous domain, each subdomain of which is homogeneous and represents a rocktype. Our first contribution is to rigorously prove convergence toward a weak solution of cell-centered finite-volume schemes with upstream mobility and without Kirchhoff's transform. Our second contribution is to numerically demonstrate the relevance of locally refining the grid at the interface between subregions, where discontinuities occur, in order to preserve an acceptable accuracy for the results computed with the schemes under consideration.

math.NA