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

• Heart rate based identification of individuals with suspected COVID-19 infection. • Semi-supervised framework using combination of auto-encoder and contrastive loss. • Contrastive convolutional auto-encoder is capable of finding proper latent attributes. • COVID-19 estimation performance declines with data shifted from symptom reported date. 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.