Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: JMIR Formative Research

Date Submitted: Jan 14, 2026
Open Peer Review Period: Jan 11, 2026 - Jan 20, 2026
Date Accepted: Sep 3, 2026
(closed for review but you can still tweet)

The final, peer-reviewed published version of this preprint can be found here:

Using Statistical Time-Series Forecasting to Predict the Resting Heart Rate From Wearable Device Data: Case Report

Chen YJ, Liu Cy, Chang YH, Lin HTE, Huang Th

Using Statistical Time-Series Forecasting to Predict the Resting Heart Rate From Wearable Device Data: Case Report

JMIR Form Res 2026;10:e91216

DOI: 10.2196/91216

PMID: 42789324

Using Statistical Time-Series Forecasting to Predict Resting Heart Rate Based on Wearable-Device Data Sources: A Case Study for a Distance Runner

  • Ying-Ju Chen; 
  • Chia-yu Liu; 
  • Ya-Han Chang; 
  • Huh-Tswen Eric Lin; 
  • Tsang-hai Huang

ABSTRACT

Background:

For endurance athletes, resting heart rate (RHR) is a well-known indicator for estimating training load status. A method for predicting the coming day's RHR would be valuable for training adjustments.

Objective:

The aim of this study is to develop a machine learning (ML) model for forecasting resting heart rate (RHR).

Methods:

Daily data (valid sample n = 624) collected from the personal wearable device of a single endurance athlete were used to establish the ML model. This ML model was built using an autoregressive integrating moving average (ARIMA) model from the Sktime package. Using heart rate as the sole data source, various strategies were employed to train the model, such as using 1-day or 7-day batch datasets and including or excluding additional features.

Results:

Models trained with all strategies showed mean forecasting values (mean = 49.371, 49.225, 49.782 & 50.101) close to observed values (mean = 49.724). However, the models trained with additional features generated a more reasonable trend (variance = 3.807 & 0.924 for 7 & 1-day batch) than those trained with only resting heart rate as the sole data source (variance = 0 for both 7 & 1-day batch).

Conclusions:

The results of the current study verified that a SARIMA model trained with sufficient features is capable of forecasting future RHR. Moreover, the present study found training the model on a weekly basis to be the best practice. Clinical Trial: N/A


 Citation

Please cite as:

Chen YJ, Liu Cy, Chang YH, Lin HTE, Huang Th

Using Statistical Time-Series Forecasting to Predict the Resting Heart Rate From Wearable Device Data: Case Report

JMIR Form Res 2026;10:e91216

DOI: 10.2196/91216

PMID: 42789324

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.