On the Risk of Misleading Reports: Diagnosing Textual Biases in Multimodal Clinical AI

• URL: http://arxiv.org/abs/2508.00171v1
• Date: Thu, 31 Jul 2025 21:35:52 GMT
• Title: On the Risk of Misleading Reports: Diagnosing Textual Biases in Multimodal Clinical AI
• Authors: David Restrepo, Ira Ktena, Maria Vakalopoulou, Stergios Christodoulidis, Enzo Ferrante,
• Abstract summary: We introduce a perturbation-based approach to quantify a model's reliance on each modality in binary classification tasks.<n>By swapping images or text between samples with opposing labels, we expose modality-specific biases.
• Score: 4.866086225040713
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Clinical decision-making relies on the integrated analysis of medical images and the associated clinical reports. While Vision-Language Models (VLMs) can offer a unified framework for such tasks, they can exhibit strong biases toward one modality, frequently overlooking critical visual cues in favor of textual information. In this work, we introduce Selective Modality Shifting (SMS), a perturbation-based approach to quantify a model's reliance on each modality in binary classification tasks. By systematically swapping images or text between samples with opposing labels, we expose modality-specific biases. We assess six open-source VLMs-four generalist models and two fine-tuned for medical data-on two medical imaging datasets with distinct modalities: MIMIC-CXR (chest X-ray) and FairVLMed (scanning laser ophthalmoscopy). By assessing model performance and the calibration of every model in both unperturbed and perturbed settings, we reveal a marked dependency on text input, which persists despite the presence of complementary visual information. We also perform a qualitative attention-based analysis which further confirms that image content is often overshadowed by text details. Our findings highlight the importance of designing and evaluating multimodal medical models that genuinely integrate visual and textual cues, rather than relying on single-modality signals.

Related papers

• Distribution-Based Masked Medical Vision-Language Model Using Structured Reports [9.306835492101413]
  Medical image-text pre-training aims to align medical images with clinically relevant text to improve model performance on various downstream tasks.<n>This work introduces an uncertainty-aware medical image-text pre-training model that enhances generalization capabilities in medical image analysis.
  arXiv  Detail & Related papers  (2025-07-29T13:31:24Z)
• Meta-Entity Driven Triplet Mining for Aligning Medical Vision-Language Models [9.76070837929117]
  Existing alignment methods prioritize separation between disease classes over segregation of fine-grained pathology attributes.<n>Here, we propose MedTrim, a novel method that enhances image-text alignment through multimodal triplet learning.<n>Our demonstrations indicate that MedTrim improves performance in downstream retrieval and classification tasks compared to state-of-the-art alignment methods.
  arXiv  Detail & Related papers  (2025-04-22T14:17:51Z)
• GCS-M3VLT: Guided Context Self-Attention based Multi-modal Medical Vision Language Transformer for Retinal Image Captioning [3.5948668755510136]
  We propose a novel vision-language model for retinal image captioning that combines visual and textual features through a guided context self-attention mechanism.<n>Experiments on the DeepEyeNet dataset demonstrate a 0.023 BLEU@4 improvement, along with significant qualitative advancements.
  arXiv  Detail & Related papers  (2024-12-23T03:49:29Z)
• ViKL: A Mammography Interpretation Framework via Multimodal Aggregation of Visual-knowledge-linguistic Features [54.37042005469384]
  We announce MVKL, the first multimodal mammography dataset encompassing multi-view images, detailed manifestations and reports. Based on this dataset, we focus on the challanging task of unsupervised pretraining. We propose ViKL, a framework that synergizes Visual, Knowledge, and Linguistic features.
  arXiv  Detail & Related papers  (2024-09-24T05:01:23Z)
• Eye-gaze Guided Multi-modal Alignment for Medical Representation Learning [65.54680361074882]
  Eye-gaze Guided Multi-modal Alignment (EGMA) framework harnesses eye-gaze data for better alignment of medical visual and textual features. We conduct downstream tasks of image classification and image-text retrieval on four medical datasets.
  arXiv  Detail & Related papers  (2024-03-19T03:59:14Z)
• C^2M-DoT: Cross-modal consistent multi-view medical report generation with domain transfer network [67.97926983664676]
  We propose a cross-modal consistent multi-view medical report generation with a domain transfer network (C2M-DoT) C2M-DoT substantially outperforms state-of-the-art baselines in all metrics.
  arXiv  Detail & Related papers  (2023-10-09T02:31:36Z)
• Learning to Exploit Temporal Structure for Biomedical Vision-Language Processing [53.89917396428747]
  Self-supervised learning in vision-language processing exploits semantic alignment between imaging and text modalities. We explicitly account for prior images and reports when available during both training and fine-tuning. Our approach, named BioViL-T, uses a CNN-Transformer hybrid multi-image encoder trained jointly with a text model.
  arXiv  Detail & Related papers  (2023-01-11T16:35:33Z)
• Multi-Granularity Cross-modal Alignment for Generalized Medical Visual Representation Learning [24.215619918283462]
  We present a novel framework for learning medical visual representations directly from paired radiology reports. Our framework harnesses the naturally exhibited semantic correspondences between medical image and radiology reports at three different levels.
  arXiv  Detail & Related papers  (2022-10-12T09:31:39Z)
• Cross-Modal Information Maximization for Medical Imaging: CMIM [62.28852442561818]
  In hospitals, data are siloed to specific information systems that make the same information available under different modalities. This offers unique opportunities to obtain and use at train-time those multiple views of the same information that might not always be available at test-time. We propose an innovative framework that makes the most of available data by learning good representations of a multi-modal input that are resilient to modality dropping at test-time.
  arXiv  Detail & Related papers  (2020-10-20T20:05:35Z)
• Semi-supervised Medical Image Classification with Relation-driven Self-ensembling Model [71.80319052891817]
  We present a relation-driven semi-supervised framework for medical image classification. It exploits the unlabeled data by encouraging the prediction consistency of given input under perturbations. Our method outperforms many state-of-the-art semi-supervised learning methods on both single-label and multi-label image classification scenarios.
  arXiv  Detail & Related papers  (2020-05-15T06:57:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.