Siamese Network for Salivary Glands Segmentation

Gabin Fodop, Aurélien Olivier, Clément Hoffmann, Ali Mansour, Sandrine Jousse-Joulin, Luc Bressollette, Benoit Clement

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review


Our project aims to help physicians and experts, to diagnose Gougerot-Sjögren’s syndrome, by segmenting salivary glands, on ultrasound images (US). Our database contains 1143 US images of left and right parotid and sub-mandibular glands obtained with a Toshiba Aplio 800 at the CHRU hospital of Brest. This manuscript proposes a method based on a Siamese architecture with a Convolutional Neural Network. To reach our goal with a relatively small database, we train the Siamese network for texture differentiation on a gray-level texture dataset, that can be transferred without any further retraining, to segment salivary glands on a US dataset.

Original languageEnglish
Title of host publicationIntelligent Decision Technologies
Subtitle of host publicationProceedings of the 14th KES-IDT 2022 Conference
EditorsIreneusz Czarnowski, Robert J. Howlett, Lakhmi C. Jain
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages9
ISBN (Electronic)978-981-19-3444-5
ISBN (Print)978-981-19-3443-8
Publication statusPublished - 2022
Externally publishedYes
Event14th International KES Conference on Intelligent Decision Technologies, KES-IDT 2022 - Virtual, Online
Duration: 20 Jun 202222 Jun 2022

Publication series

NameSmart Innovation, Systems and Technologies
ISSN (Print)2190-3018
ISSN (Electronic)2190-3026


Conference14th International KES Conference on Intelligent Decision Technologies, KES-IDT 2022
CityVirtual, Online


  • Convolutional neural network
  • Gougerot-Sjögren’s syndrome
  • Siamese network
  • Ultrasound imaging


Dive into the research topics of 'Siamese Network for Salivary Glands Segmentation'. Together they form a unique fingerprint.

Cite this