MasHeNe: A Benchmark for Head and Neck CT Mass Segmentation using Window-Enhanced Mamba with Frequency-Domain Integration

2025-12-01

AI Summary

To address the lack of publicly available segmentation benchmarks for head-and-neck space-occupying lesions—particularly cysts and tumors—this work introduces MasHeNe, the first annotated contrast-enhanced CT dataset (3,779 axial slices) covering both benign and malignant lesions. We further propose WEMF, a U-shaped network incorporating triple-window CT preprocessing, frequency-domain attention, and a Mamba-based state-space backbone, with optimized skip connections to enhance robustness in segmenting heterogeneous lesions. On MasHeNe, WEMF achieves Dice 70.45%, IoU 66.89%, NSD 72.33%, and HD95 5.12 mm—significantly outperforming existing methods. This study establishes the first fine-grained segmentation benchmark for head-and-neck masses and empirically validates the efficacy of frequency-domain modeling and state-space architectures in medical image segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web

Abstract

Head and neck masses are space-occupying lesions that can compress the airway and esophagus and may affect nerves and blood vessels. Available public datasets primarily focus on malignant lesions and often overlook other space-occupying conditions in this region. To address this gap, we introduce MasHeNe, an initial dataset of 3,779 contrast-enhanced CT slices that includes both tumors and cysts with pixel-level annotations. We also establish a benchmark using standard segmentation baselines and report common metrics to enable fair comparison. In addition, we propose the Windowing-Enhanced Mamba with Frequency integration (WEMF) model. WEMF applies tri-window enhancement to enrich the input appearance before feature extraction. It further uses multi-frequency attention to fuse information across skip connections within a U-shaped Mamba backbone. On MasHeNe, WEMF attains the best performance among evaluated methods, with a Dice of 70.45%, IoU of 66.89%, NSD of 72.33%, and HD95 of 5.12 mm. This model indicates stable and strong results on this challenging task. MasHeNe provides a benchmark for head-and-neck mass segmentation beyond malignancy-only datasets. The observed error patterns also suggest that this task remains challenging and requires further research. Our dataset and code are available at https://github.com/drthaodao3101/MasHeNe.git.

Problem

• Develops a dataset for head and neck mass segmentation beyond only malignant lesions
• Proposes a model with window enhancement and frequency-domain integration for segmentation
• Establishes a benchmark to enable fair comparison on this challenging task

Innovation

• Tri-window enhancement enriches input appearance
• Multi-frequency attention fuses skip connection information
• U-shaped Mamba backbone integrates frequency-domain features