Salvato in:
Dettagli Bibliografici
Autori principali: Bader, Felicia, Seeböck, Philipp, Bartashova, Anastasia, Attenberger, Ulrike, Langs, Georg
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2604.13970
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911595876057088
author Bader, Felicia
Seeböck, Philipp
Bartashova, Anastasia
Attenberger, Ulrike
Langs, Georg
author_facet Bader, Felicia
Seeböck, Philipp
Bartashova, Anastasia
Attenberger, Ulrike
Langs, Georg
contents In diagnostic reports, experts encode complex imaging data into clinically actionable information. They describe subtle pathological findings that are meaningful in their anatomical context. Reports follow relatively consistent structures, expressing diagnostic information with few words that are often associated with tiny but consequential image observations. Standard vision language models struggle to identify the associations between these informative text components and small locations in the images. Here, we propose "MApLe", a multi-task, multi-instance vision language alignment approach that overcomes these limitations. It disentangles the concepts of anatomical region and diagnostic finding, and links local image information to sentences in a patch-wise approach. Our method consists of a text embedding trained to capture anatomical and diagnostic concepts in sentences, a patch-wise image encoder conditioned on anatomical structures, and a multi-instance alignment of these representations. We demonstrate that MApLe can successfully align different image regions and multiple diagnostic findings in free-text reports. We show that our model improves the alignment performance compared to state-of-the-art baseline models when evaluated on several downstream tasks. The code is available at https://github.com/cirmuw/MApLe.
format Preprint
id arxiv_https___arxiv_org_abs_2604_13970
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MApLe: Multi-instance Alignment of Diagnostic Reports and Large Medical Images
Bader, Felicia
Seeböck, Philipp
Bartashova, Anastasia
Attenberger, Ulrike
Langs, Georg
Computer Vision and Pattern Recognition
In diagnostic reports, experts encode complex imaging data into clinically actionable information. They describe subtle pathological findings that are meaningful in their anatomical context. Reports follow relatively consistent structures, expressing diagnostic information with few words that are often associated with tiny but consequential image observations. Standard vision language models struggle to identify the associations between these informative text components and small locations in the images. Here, we propose "MApLe", a multi-task, multi-instance vision language alignment approach that overcomes these limitations. It disentangles the concepts of anatomical region and diagnostic finding, and links local image information to sentences in a patch-wise approach. Our method consists of a text embedding trained to capture anatomical and diagnostic concepts in sentences, a patch-wise image encoder conditioned on anatomical structures, and a multi-instance alignment of these representations. We demonstrate that MApLe can successfully align different image regions and multiple diagnostic findings in free-text reports. We show that our model improves the alignment performance compared to state-of-the-art baseline models when evaluated on several downstream tasks. The code is available at https://github.com/cirmuw/MApLe.
title MApLe: Multi-instance Alignment of Diagnostic Reports and Large Medical Images
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.13970