NEURAL: Attention-Guided Pruning for Unified Multimodal Resource-Constrained Clinical Evaluation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Joshi, Devvrat, Rekik, Islem
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913988424499200
author Joshi, Devvrat
Rekik, Islem
author_facet Joshi, Devvrat
Rekik, Islem
contents The rapid growth of multimodal medical imaging data presents significant storage and transmission challenges, particularly in resource-constrained clinical settings. We propose NEURAL, a novel framework that addresses this by using semantics-guided data compression. Our approach repurposes cross-attention scores between the image and its radiological report from a fine-tuned generative vision-language model to structurally prune chest X-rays, preserving only diagnostically critical regions. This process transforms the image into a highly compressed, graph representation. This unified graph-based representation fuses the pruned visual graph with a knowledge graph derived from the clinical report, creating a universal data structure that simplifies downstream modeling. Validated on the MIMIC-CXR and CheXpert Plus dataset for pneumonia detection, NEURAL achieves a 93.4-97.7\% reduction in image data size while maintaining a high diagnostic performance of 0.88-0.95 AUC, outperforming other baseline models that use uncompressed data. By creating a persistent, task-agnostic data asset, NEURAL resolves the trade-off between data size and clinical utility, enabling efficient workflows and teleradiology without sacrificing performance. Our NEURAL code is available at https://github.com/basiralab/NEURAL.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09715
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NEURAL: Attention-Guided Pruning for Unified Multimodal Resource-Constrained Clinical Evaluation
Joshi, Devvrat
Rekik, Islem
Computer Vision and Pattern Recognition
Machine Learning
The rapid growth of multimodal medical imaging data presents significant storage and transmission challenges, particularly in resource-constrained clinical settings. We propose NEURAL, a novel framework that addresses this by using semantics-guided data compression. Our approach repurposes cross-attention scores between the image and its radiological report from a fine-tuned generative vision-language model to structurally prune chest X-rays, preserving only diagnostically critical regions. This process transforms the image into a highly compressed, graph representation. This unified graph-based representation fuses the pruned visual graph with a knowledge graph derived from the clinical report, creating a universal data structure that simplifies downstream modeling. Validated on the MIMIC-CXR and CheXpert Plus dataset for pneumonia detection, NEURAL achieves a 93.4-97.7\% reduction in image data size while maintaining a high diagnostic performance of 0.88-0.95 AUC, outperforming other baseline models that use uncompressed data. By creating a persistent, task-agnostic data asset, NEURAL resolves the trade-off between data size and clinical utility, enabling efficient workflows and teleradiology without sacrificing performance. Our NEURAL code is available at https://github.com/basiralab/NEURAL.
title NEURAL: Attention-Guided Pruning for Unified Multimodal Resource-Constrained Clinical Evaluation
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2508.09715