Automated detection of diabetic subject using pre-trained 2D-CNN models with frequency spectrum images extracted from heart rate signals

Ozal Yildirim*, Muhammed Talo, Betul Ay, Ulas Baran Baloglu, Galip Aydin, U. Rajendra Acharya

*Corresponding author for this work

Research output: Contribution to journalArticle (Academic Journal)peer-review

33 Citations (Scopus)
77 Downloads (Pure)

Abstract

In this study, a deep-transfer learning approach is proposed for the automated diagnosis of diabetes mellitus (DM), using heart rate (HR) signals obtained from electrocardiogram (ECG) data. Recent progress in deep learning has contributed significantly to improvement in the quality of healthcare. In order for deep learning models to perform well, large datasets are required for training. However, a difficulty in the biomedical field is the lack of clinical data with expert annotation. A recent, commonly implemented technique to train deep learning models using small datasets is to transfer the weighting, developed from a large dataset, to the current model. This deep learning transfer strategy is generally employed for two-dimensional signals. Herein, the weighting of models pre-trained using two-dimensional large image data was applied to one-dimensional HR signals. The one-dimensional HR signals were then converted into frequency spectrum images, which were utilized for application to well-known pre-trained models, specifically: AlexNet, VggNet, ResNet, and DenseNet. The DenseNet pre-trained model yielded the highest classification average accuracy of 97.62%, and sensitivity of 100%, to detect DM subjects via HR signal recordings. In the future, we intend to further test this developed model by utilizing additional data along with cloud-based storage to diagnose DM via heart signal analysis.

Original languageEnglish
Article number103387
Number of pages10
JournalComputers in Biology and Medicine
Volume113
Early online date9 Aug 2019
DOIs
Publication statusPublished - 1 Oct 2019

Keywords

  • Deep learning
  • Diabetes mellitus
  • Heart rate signals
  • Transfer learning

Fingerprint Dive into the research topics of 'Automated detection of diabetic subject using pre-trained 2D-CNN models with frequency spectrum images extracted from heart rate signals'. Together they form a unique fingerprint.

Cite this