arXiv:2512.07191v1 Announce Type: new 
Abstract: Medical image segmentation remains challenging due to intensity inhomogeneity, noise, blurred boundaries, and irregular structures. Traditional level set methods, while effective in certain cases, often depend on approximate bias field estimations and therefore struggle under severe non-uniform imaging conditions. To address these limitations, we propose a novel variational Reflectance-based Level Set Model (RefLSM), which explicitly integrates Retinex-inspired reflectance decomposition into the segmentation framework. By decomposing the observed image into reflectance and bias field components, RefLSM directly segments the reflectance, which is invariant to illumination and preserves fine structural details. Building on this foundation, we introduce two key innovations for enhanced precision and robustness. First, a linear structural prior steers the smoothed reflectance gradients toward a data-driven reference, providing reliable geometric guidance in noisy or low-contrast scenes. Second, a relaxed binary level-set is embedded in RefLSM and enforced via convex relaxation and sign projection, yielding stable evolution and avoiding reinitialization-induced diffusion. The resulting variational problem is solved efficiently using an ADMM-based optimization scheme. Extensive experiments on multiple medical imaging datasets demonstrate that RefLSM achieves superior segmentation accuracy, robustness, and computational efficiency compared to state-of-the-art level set methods.

تم اقتراح نموذج جديد قائم على مستوى الانعكاس (RefLSM) لتحسين تقسيم الصور الطبية وتصحيح مجالات التحيز، حيث يتناول التحديات مثل عدم تجانس الكثافة والضوضاء. يدمج هذا النموذج تحليل الانعكاس المستوحى من Retinex، مما يسمح بالتقسيم المباشر للانعكاس مع الحفاظ على التفاصيل الهيكلية.

Se ha propuesto un nuevo modelo de segmentación de imágenes médicas basado en un nivel de reflexión (RefLSM) para mejorar la segmentación y la corrección de campos de sesgo, abordando desafíos como la inhomogeneidad de intensidad y el ruido. Este modelo integra una descomposición de reflexión inspirada en Retinex, permitiendo la segmentación directa de la reflexión mientras se preservan los detalles estructurales.

Un nouveau modèle de segmentation d'image médicale basé sur un niveau de réflexion (RefLSM) a été proposé pour améliorer la segmentation et la correction des champs de biais, en répondant à des défis tels que l'hétérogénéité d'intensité et le bruit. Ce modèle intègre une décomposition de réflexion inspirée de Retinex, permettant une segmentation directe de la réflexion tout en préservant les détails structurels.

A new variational Reflectance-based Level Set Model (RefLSM) has been proposed to improve medical image segmentation and bias-field correction, addressing challenges such as intensity inhomogeneity and noise. This model integrates Retinex-inspired reflectance decomposition, allowing for direct segmentation of reflectance while preserving structural details.

RefLSM: Linearized Structural-Prior Reflectance Model for Medical Image Segmentation and Bias-Field Correction

Was this article worth reading? Share it

LucidQuery AI

LexiStock AI

Resub

Lenso.ai

FiltrixAI

Resolve

Ready to build your own newsroom?