Skip to main navigation Skip to search Skip to main content

Automatic Diagnosis of Pectus Excavatum from CT Images Using a Joint CNN-LSTM Model

  • Yizhi Liao
  • , Haiyu Zhou
  • , Longhan Xie
  • , Siqi Cai*
  • *Corresponding author for this work
  • School of Computing and Information Systems
  • Division of Thoracic Surgery
  • South China University of Technology
  • National University of Singapore

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

Abstract

Pectus excavatum (PE) is one of the most common congenital sternal deformities. Accurate preoperative diagnosis of PE is of great significance for subsequent correction and improvement of the patient's quality of life. However, current diagnostic methods rely on the calculation of some PE indices, which is a heavy workload for physical therapists and suffers from measurement errors. To address this issue, we propose an end-to-end automatic assessment of PE, which features a cascaded structure of CNN and LSTM. Specifically, the high-level feature representations of CT images are extracted by the pretrained CNN and then processed through the LSTM layers for classification. In addition, we build up a medical image dataset for PE diagnosis by collecting chest CT images of 42 subjects. Results on this dataset show that the proposed CNN-LSTM framework achieves a relatively high accuracy of 90.20%, which provides a new perspective for the automatic diagnosis of PE in clinics.

Original languageEnglish
Title of host publication9th 2023 International Conference on Control, Decision and Information Technologies, CoDIT 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages853-857
Number of pages5
ISBN (Electronic)9798350311402
DOIs
StatePublished - 2023
Externally publishedYes
Event9th International Conference on Control, Decision and Information Technologies, CoDIT 2023 - Rome, Italy
Duration: 3 Jul 20236 Jul 2023

Publication series

Name9th 2023 International Conference on Control, Decision and Information Technologies, CoDIT 2023

Conference

Conference9th International Conference on Control, Decision and Information Technologies, CoDIT 2023
Country/TerritoryItaly
CityRome
Period3/07/236/07/23

Fingerprint

Dive into the research topics of 'Automatic Diagnosis of Pectus Excavatum from CT Images Using a Joint CNN-LSTM Model'. Together they form a unique fingerprint.

Cite this