Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Toward Content-based Indexing and Retrieval of Head and Neck CT with Abscess Segmentation

Thao Thi Phuong Dao, Tan-Cong Nguyen, Trong-Le Do, Truong Hoang Viet, Nguyen Chi Thanh, Huynh Nguyen Thuan, Do Vo Cong Nguyen, Minh-Khoi Pham, Mai-Khiem Tran, Viet-Tham Huynh, Trong-Thuan Nguyen, Trung-Nghia Le, Vo Thanh Toan, Tam V. Nguyen, Minh-Triet Tran, Thanh Dinh Le

TL;DR

<3-5 sentence high-level summary> The paper tackles the clinical need for accurate segmentation and retrieval of head and neck abscesses on contrast-enhanced CT. It introduces AbscessHeNe, a dataset with 4,926 annotated CT slices from 47 patients, linking pixel-wise masks to structured clinical metadata to support segmentation and content-based indexing. A comprehensive benchmark across CNN, Transformer, and Mamba-based architectures, with VM-UNet achieving the best performance (DSC 0.39, mIoU 0.27, NSD 0.67, HD95 21.27 mm), reveals substantial boundary delineation challenges. Beyond segmentation, the work outlines a retrieval prototype enabling query-by-image/mask and metadata-guided case search, laying the groundwork for retrieval-augmented clinical decision support in head-and-neck imaging.

Abstract

Abscesses in the head and neck represent an acute infectious process that can potentially lead to sepsis or mortality if not diagnosed and managed promptly. Accurate detection and delineation of these lesions on imaging are essential for diagnosis, treatment planning, and surgical intervention. In this study, we introduce AbscessHeNe, a curated and comprehensively annotated dataset comprising 4,926 contrast-enhanced CT slices with clinically confirmed head and neck abscesses. The dataset is designed to facilitate the development of robust semantic segmentation models that can accurately delineate abscess boundaries and evaluate deep neck space involvement, thereby supporting informed clinical decision-making. To establish performance baselines, we evaluate several state-of-the-art segmentation architectures, including CNN, Transformer, and Mamba-based models. The highest-performing model achieved a Dice Similarity Coefficient of 0.39, Intersection-over-Union of 0.27, and Normalized Surface Distance of 0.67, indicating the challenges of this task and the need for further research. Beyond segmentation, AbscessHeNe is structured for future applications in content-based multimedia indexing and case-based retrieval. Each CT scan is linked with pixel-level annotations and clinical metadata, providing a foundation for building intelligent retrieval systems and supporting knowledge-driven clinical workflows. The dataset will be made publicly available at https://github.com/drthaodao3101/AbscessHeNe.git.

Toward Content-based Indexing and Retrieval of Head and Neck CT with Abscess Segmentation

TL;DR

<3-5 sentence high-level summary> The paper tackles the clinical need for accurate segmentation and retrieval of head and neck abscesses on contrast-enhanced CT. It introduces AbscessHeNe, a dataset with 4,926 annotated CT slices from 47 patients, linking pixel-wise masks to structured clinical metadata to support segmentation and content-based indexing. A comprehensive benchmark across CNN, Transformer, and Mamba-based architectures, with VM-UNet achieving the best performance (DSC 0.39, mIoU 0.27, NSD 0.67, HD95 21.27 mm), reveals substantial boundary delineation challenges. Beyond segmentation, the work outlines a retrieval prototype enabling query-by-image/mask and metadata-guided case search, laying the groundwork for retrieval-augmented clinical decision support in head-and-neck imaging.

Abstract

Abscesses in the head and neck represent an acute infectious process that can potentially lead to sepsis or mortality if not diagnosed and managed promptly. Accurate detection and delineation of these lesions on imaging are essential for diagnosis, treatment planning, and surgical intervention. In this study, we introduce AbscessHeNe, a curated and comprehensively annotated dataset comprising 4,926 contrast-enhanced CT slices with clinically confirmed head and neck abscesses. The dataset is designed to facilitate the development of robust semantic segmentation models that can accurately delineate abscess boundaries and evaluate deep neck space involvement, thereby supporting informed clinical decision-making. To establish performance baselines, we evaluate several state-of-the-art segmentation architectures, including CNN, Transformer, and Mamba-based models. The highest-performing model achieved a Dice Similarity Coefficient of 0.39, Intersection-over-Union of 0.27, and Normalized Surface Distance of 0.67, indicating the challenges of this task and the need for further research. Beyond segmentation, AbscessHeNe is structured for future applications in content-based multimedia indexing and case-based retrieval. Each CT scan is linked with pixel-level annotations and clinical metadata, providing a foundation for building intelligent retrieval systems and supporting knowledge-driven clinical workflows. The dataset will be made publicly available at https://github.com/drthaodao3101/AbscessHeNe.git.

Paper Structure

This paper contains 18 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related works
  3. Medical Image Segmentation and Retrieval
  4. Head and Neck CT Scans Datasets for Segmentation and Retrieval
  5. AbscessHeNe Dataset
  6. Clinical Definition
  7. Dataset Construction
  8. Data Collection
  9. Data Annotation
  10. Benchmark Suite
  11. Segmentation Baselines
  12. Quantitative Comparison
  13. Qualitative Comparison
  14. Retrieval Prototype
  15. Discussion
  16. ...and 3 more sections

Figures (4)

  • Figure 1: Comprehensive pipeline for building AbscessHeNe dataset.
  • Figure 2: A sample of AbscessHeNe. Left-hand: original image and Right-hand: Image with mask (green)
  • Figure 3: Visualization of results from the different models on our AbscessHeNe dataset for segmentation.
  • Figure 4: Conceptual prototype for retrieval with AbscessHeNe, illustrating how query images and metadata can be processed and matched to return clinically similar cases.