Skip to main navigation Skip to search Skip to main content

DiENTeS: Dynamic ENTity Segmentation with Local-Global Transformers

  • Institute of Machine Learning in Biomedical Imaging
  • Technical University of Munich
  • King's College London

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Semantic segmentation is crucial for accurately identifying anatomical structures and pathological anomalies in medical images, playing a vital role in diagnostics, treatment planning, and disease progression monitoring. Despite significant advancements, the development of flexible and generalizable algorithms that can adapt to the diverse shapes, sizes, and textures of various anatomical regions remains challenging. In this work, we introduce the Dynamic ENTity Segmentation (DiENTeS) model, which leverages Local-global Transformers for 3D medical segmentation. Our model utilizes a transformer-based backbone to extract localized features and propagate them to form a comprehensive global representation. Additionally, we incorporate language features to guide the segmentation process, enabling the generation of specialized convolutional kernels for each category. This approach allows DiENTeS to tackle semantic segmentation as a class-agnostic entity segmentation problem. We validate our method using the ToothFairy2 Challenge, demonstrating its effectiveness in segmenting multiple structures in the maxillofacial region. We will make our code and models publicly available.

Original languageEnglish
Title of host publicationSupervised and Semi-supervised Multi-structure Segmentation and Landmark Detection in Dental Data - MICCAI 2024 Challenges
Subtitle of host publicationToothFairy 2024, 3DTeethLand 2024, and STS 2024, Held in Conjunction with MICCAI 2024, Proceedings
EditorsYaqi Wang, Dahong Qian, Shuai Wang, Achraf Ben-Hamadou, Sergi Pujades, Luca Lumetti, Federico Bolelli, Costantino Grana
PublisherSpringer Science and Business Media Deutschland GmbH
Pages21-29
Number of pages9
ISBN (Print)9783031889769
DOIs
StatePublished - 2025
Event2nd International ToothFairy2: Multi-Structure Segmentation in CBCT Volumes, ToothFairy 2024, 3D Teeth Landmarks Detection Challenge, 3DTeethLand 2024, Semi-supervised Teeth Segmentation, STS 2024 held in conjunction with the 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2024 - Marrakesh, Morocco
Duration: 6 Oct 20246 Oct 2024

Publication series

NameLecture Notes in Computer Science
Volume15571 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International ToothFairy2: Multi-Structure Segmentation in CBCT Volumes, ToothFairy 2024, 3D Teeth Landmarks Detection Challenge, 3DTeethLand 2024, Semi-supervised Teeth Segmentation, STS 2024 held in conjunction with the 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2024
Country/TerritoryMorocco
CityMarrakesh
Period6/10/246/10/24

Keywords

  • Semantic segmentation
  • ToothFairy2 Challenge
  • multimodal segmentation
  • vision transformers

Fingerprint

Dive into the research topics of 'DiENTeS: Dynamic ENTity Segmentation with Local-Global Transformers'. Together they form a unique fingerprint.

Cite this