Cancer-related fatigue classification based on heart rate variability signals from wearables

Chi-Huang Shih; Pai-Chien Chou; Jin-Hua Chen; Ting-Ling Chou; Jun-Hung Lai; Chi-Yu Lu; Tsai-Wei Huang

doi:10.3389/fmed.2023.1103979

Cancer-related fatigue classification based on heart rate variability signals from wearables

Front Med (Lausanne). 2023 Apr 26:10:1103979. doi: 10.3389/fmed.2023.1103979. eCollection 2023.

Authors

Chi-Huang Shih¹, Pai-Chien Chou^{2

3}, Jin-Hua Chen⁴, Ting-Ling Chou⁵, Jun-Hung Lai⁶, Chi-Yu Lu⁷, Tsai-Wei Huang^{5

7

8

9}

Affiliations

¹ Department of Computer Science and Information Engineering, National Chin-Yi University of Technology, Taichung, Taiwan.
² Division of Pulmonary Medicine, Department of Internal Medicine, Taipei Medical University Hospital, Taipei, Taiwan.
³ Division of Thoracic Medicine, Department of Internal Medicine, School of Medicine, Taipei Medical University, Taipei, Taiwan.
⁴ Graduate Institute of Data Science, Taipei Medical University, Taipei, Taiwan.
⁵ School of Nursing, College of Nursing, Taipei Medical University, Taipei, Taiwan.
⁶ Division of Gastroenterology and Hepatology, Department of Internal Medicine, Erlin Christian Hospital, Changhua, Taiwan.
⁷ Department of Nursing, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan.
⁸ Cochrane Taiwan, Taipei Medical University, Taipei, Taiwan.
⁹ Research Center in Nursing Clinical Practice, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan.

Abstract

Background: Cancer-related fatigue (CRF) is the most distressing side effect in cancer patients and affects the survival rate. However, most patients do not report their fatigue level. This study is aimed to develop an objective CRF assessment method based on heart rate variability (HRV).

Methods: In this study, patients with lung cancer who received chemotherapy or target therapy were enrolled. Patients wore wearable devices with photoplethysmography that regularly recorded HRV parameters for seven consecutive days and completed the Brief Fatigue Inventory (BFI) questionnaire. The collected parameters were divided into the active and sleep phase parameters to allow tracking of fatigue variation. Statistical analysis was used to identify correlations between fatigue scores and HRV parameters.

Findings: In this study, 60 patients with lung cancer were enrolled. The HRV parameters including the low-frequency/high-frequency (LF/HF) ratio and the LF/HF disorder ratio in the active phase and the sleep phase were extracted. A linear classifier with HRV-based cutoff points achieved correct classification rates of 73 and 88% for mild and moderate fatigue levels, respectively.

Conclusion: Fatigue was effectively identified, and the data were effectively classified using a 24-h HRV device. This objective fatigue monitoring method may enable clinicians to effectively handle fatigue problems.

Keywords: LF/HF ratio; cancer-related fatigue; heart rate variability; photoplethysmography; wearables.

Grants and funding

The funding for this study was provided by the National Science and Technology Council of Taiwan (grants MOST 110-2622-E-038-006 and NSTC 111-2622-E-038-003).