arXiv:2511.00477v1 Announce Type: cross 
Abstract: Algorithmic bias in medical imaging can perpetuate health disparities, yet its causes remain poorly understood in segmentation tasks. While fairness has been extensively studied in classification, segmentation remains underexplored despite its clinical importance. In breast cancer segmentation, models exhibit significant performance disparities against younger patients, commonly attributed to physiological differences in breast density. We audit the MAMA-MIA dataset, establishing a quantitative baseline of age-related bias in its automated labels, and reveal a critical Biased Ruler effect where systematically flawed labels for validation misrepresent a model's actual bias. However, whether this bias originates from lower-quality annotations (label bias) or from fundamentally more challenging image characteristics remains unclear. Through controlled experiments, we systematically refute hypotheses that the bias stems from label quality sensitivity or quantitative case difficulty imbalance. Balancing training data by difficulty fails to mitigate the disparity, revealing that younger patient cases are intrinsically harder to learn. We provide direct evidence that systemic bias is learned and amplified when training on biased, machine-generated labels, a critical finding for automated annotation pipelines. This work introduces a systematic framework for diagnosing algorithmic bias in medical segmentation and demonstrates that achieving fairness requires addressing qualitative distributional differences rather than merely balancing case counts.

تدرس دراسة حديثة التحيز الخوارزمي الموجود في التصوير الطبي، وخاصة في مهام التقسيم المتعلقة بسرطان الثدي. وتبرز كيف يواجه المرضى الأصغر سناً غالباً تفاوتات في الأداء بسبب الاختلافات الفسيولوجية في كثافة الثدي. إن فهم هذه التحيزات أمر بالغ الأهمية، حيث يمكن أن يساعد في معالجة الفجوات الصحية وتحسين النتائج السريرية لجميع المرضى، مما يضمن أن تكون التقنيات الطبية عادلة وفعالة.

Un estudio reciente investiga el sesgo algorítmico presente en la imagenología médica, especialmente en las tareas de segmentación relacionadas con el cáncer de mama. Destaca cómo los pacientes más jóvenes a menudo enfrentan disparidades en el rendimiento debido a diferencias fisiológicas en la densidad mamaria. Comprender estos sesgos es crucial, ya que puede ayudar a abordar las disparidades en salud y mejorar los resultados clínicos para todos los pacientes, asegurando que las tecnologías médicas sean justas y efectivas.

Une étude récente examine le biais algorithmique présent dans l'imagerie médicale, en particulier dans les tâches de segmentation liées au cancer du sein. Elle souligne comment les patients plus jeunes font souvent face à des disparités de performance en raison de différences physiologiques dans la densité mammaire. Comprendre ces biais est crucial car cela peut aider à résoudre les disparités en matière de santé et à améliorer les résultats cliniques pour tous les patients, garantissant que les technologies médicales soient justes et efficaces.

A recent study investigates the algorithmic bias present in medical imaging, particularly in segmentation tasks related to breast cancer. It highlights how younger patients often face performance disparities due to physiological differences in breast density. Understanding these biases is crucial as it can help address health disparities and improve clinical outcomes for all patients, ensuring that medical technologies are fair and effective.

Investigating Label Bias and Representational Sources of Age-Related Disparities in Medical Segmentation

arXiv:2601.08336v1 Announce Type: new 
Abstract: Automatic integration of whole slide images (WSIs) and gene expression profiles has demonstrated substantial potential in precision clinical diagnosis and cancer progression studies. However, most existing studies focus on individual gene sequences and slide level classification tasks, with limited attention to spatial transcriptomics and patch level applications. To address this limitation, we propose a multimodal network, BioMorphNet, which automatically integrates tissue morphological features and spatial gene expression to support tissue classification and differential gene analysis. For considering morphological features, BioMorphNet constructs a graph to model the relationships between target patches and their neighbors, and adjusts the response strength based on morphological and molecular level similarity, to better characterize the tumor microenvironment. In terms of multimodal interactions, BioMorphNet derives clinical pathway features from spatial transcriptomic data based on a predefined pathway database, serving as a bridge between tissue morphology and gene expression. In addition, a novel learnable pathway module is designed to automatically simulate the biological pathway formation process, providing a complementary representation to existing clinical pathways. Compared with the latest morphology gene multimodal methods, BioMorphNet's average classification metrics improve by 2.67%, 5.48%, and 6.29% for prostate cancer, colorectal cancer, and breast cancer datasets, respectively. BioMorphNet not only classifies tissue categories within WSIs accurately to support tumor localization, but also analyzes differential gene expression between tissue categories based on prediction confidence, contributing to the discovery of potential tumor biomarkers.

تم اقتراح شبكة متعددة الوسائط جديدة تُدعى BioMorphNet لتحسين تصنيف الأنسجة وتحليل صور الشرائح الكاملة من خلال دمج الميزات الشكلية والتعبير الجيني المكاني، مما يعالج القيود في منهجيات البحث الحالية حول السرطان.

Se ha propuesto una nueva red multimodal llamada BioMorphNet para mejorar la clasificación de tejidos y el análisis de imágenes de diapositivas completas mediante la integración de características morfológicas y expresión génica espacial, abordando las limitaciones de las metodologías actuales en la investigación del cáncer.

Un nouveau réseau multimodal nommé BioMorphNet a été proposé pour améliorer la classification des tissus et l'analyse des images à grande échelle en intégrant des caractéristiques morphologiques et l'expression génique spatiale, répondant aux limitations des méthodologies de recherche sur le cancer actuelles.

A new multimodal network named BioMorphNet has been proposed to enhance tissue classification and whole-slide image analysis by integrating morphological features and spatial gene expression, addressing limitations in current cancer research methodologies.

Tissue Classification and Whole-Slide Images Analysis via Modeling of the Tumor Microenvironment and Biological Pathways

One More Thing in AI – Your Shortcut to AI Mastery

Investigating Label Bias and Representational Sources of Age-Related Disparities in Medical Segmentation

Was this article worth reading? Share it

One More Thing in AI

Emma: AI Food Scanner

LucidQuery AI

Meteoria

Twofold Health

AIPortalX

Ready to build your own newsroom?