Fitbeat: COVID-19 estimation based on wristband heart rate using a contrastive convolutional auto-encoder

Shuo Liu*, Jing Han, Estela Laporta Puyal, Spyridon Kontaxis, Shaoxiong Sun, Patrick Locatelli, Judith Dineley, Florian B. Pokorny, Gloria Dalla Costa, Letizia Leocani, Ana Isabel Guerrero, Carlos Nos, Ana Zabalza, Per Soelberg Sørensen, Mathias Buron, Melinda Magyari, Yatharth Ranjan, Zulqarnain Rashid, Pauline Conde, Callum StewartAmos A. Folarin, Richard JB Dobson, Raquel Bailón, Srinivasan Vairavan, Nicholas Cummins, Vaibhav A. Narayan, Matthew Hotopf, Giancarlo Comi, Björn Schuller, RADAR A.D.A.R.C.N.S. Consortium

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

24 Citations (Scopus)

Abstract

This study proposes a contrastive convolutional auto-encoder (contrastive CAE), a combined architecture of an auto-encoder and contrastive loss, to identify individuals with suspected COVID-19 infection using heart-rate data from participants with multiple sclerosis (MS) in the ongoing RADAR-CNS mHealth research project. Heart-rate data was remotely collected using a Fitbit wristband. COVID-19 infection was either confirmed through a positive swab test, or inferred through a self-reported set of recognised symptoms of the virus. The contrastive CAE outperforms a conventional convolutional neural network (CNN), a long short-term memory (LSTM) model, and a convolutional auto-encoder without contrastive loss (CAE). On a test set of 19 participants with MS with reported symptoms of COVID-19, each one paired with a participant with MS with no COVID-19 symptoms, the contrastive CAE achieves an unweighted average recall of 95.3%, a sensitivity of 100% and a specificity of 90.6%, an area under the receiver operating characteristic curve (AUC-ROC) of 0.944, indicating a maximum successful detection of symptoms in the given heart rate measurement period, whilst at the same time keeping a low false alarm rate.

Original languageEnglish
Article number108403
JournalPATTERN RECOGNITION
Volume123
Early online date26 Oct 2021
DOIs
Publication statusPublished - Mar 2022

Keywords

  • Anomaly detection
  • Contrastive learning
  • Convolutional auto-encoder
  • COVID-19
  • Respiratory tract infection

Fingerprint

Dive into the research topics of 'Fitbeat: COVID-19 estimation based on wristband heart rate using a contrastive convolutional auto-encoder'. Together they form a unique fingerprint.

Cite this