Multi-scale Contrastive Learning for Gastroenteroscopy Classification

Dan Li, Xuechen Li, Zhibin Peng, Wenting Chen, Linlin Shen, Guangyao Wu

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


In gastroenteroscopy image analysis, numerous CADs demonstrate that deep learning aids doctors' diagnosis. The shapes and sizes of the lesions are varied. And in the clinic, the dataset appears to be data imbalanced. However, existing methods directly classify by texture and ignore lesions with various shapes and sizes. To address the issue above, we propose a deep neural network, which consists of multi-scale feature extraction, contrastive feature learning and a multi-scale feature fusion module. We train the contrastive feature learning module and multi-scale feature fusion module simultaneously to alleviate the issue of data distribution differences. Thus, the proposed network can better identify various categories. Extensive experiments on the Hyper Kvasir dataset show that the proposed Hybrid-M2CL outperforms the benchmark proposed by the dataset with 5.0% Macro Precision, 3.3% Macro Recall, 3.4% Macro F1-score, 3.3% Micro Precision, 3.6% MCC. In addition, it outperforms the SOTA by 1.1% Macro F1-score, 2.6% MCC, and 2.0% B-ACC.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE 36th International Symposium on Computer-Based Medical Systems, CBMS 2023
EditorsRosa Sicilia, Bridget Kane, Joao Rafael Almeida, Myra Spiliopoulou, Jose Alberto Benitez Andrades, Giuseppe Placidi, Alejandro Rodriguez Gonzalez
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages7
ISBN (Electronic)9798350312249
Publication statusPublished - 2023
Externally publishedYes
Event36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023 - L�Aquila, Italy
Duration: 22 Jun 202324 Jun 2023

Publication series

NameProceedings - IEEE Symposium on Computer-Based Medical Systems
ISSN (Print)1063-7125


Conference36th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2023


  • contrastive learning
  • gastroenterology classification
  • multi-scale
  • supervised learning

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications


Dive into the research topics of 'Multi-scale Contrastive Learning for Gastroenteroscopy Classification'. Together they form a unique fingerprint.

Cite this