Yip J, Grießhammer S, Leutheuser H, Richer R, Lu H, Kölpin A, Eskofier B, Ostgathe C, Steigleder T (2025)
Publication Status: Submitted
Publication Type: Unpublished / Preprint
Future Publication Type: Journal article
Publication year: 2025
DOI: 10.36227/techrxiv.174495693.37175254/v1
In palliative and end-of-life care, effective communication about expected death is critical for patient management and family support. This study explores the feasibility of burden-free radar data to distinguish between alive and deceased patients at such a vulnerable time. Our research also aims to identify the important signal features for this distinction. To achieve these goals, we collected and annotated a dataset of radar recordings from 16 palliative care patients in the dying phase. The radar data can be used for motion biomarkers to predict the vital status of patients. We applied several machine learning algorithms to detect underlying patterns in the data. Quantitative evaluation of these algorithms yielded balanced accuracy rates ranging from 92% to 98%. In addition, qualitative insights from medical professionals specializing in end-of-life care were instrumental in validating and interpreting the results. This interdisciplinary research highlights the potential of radar technology as a non-invasive tool for monitoring vital status in palliative care, providing valuable insights into patient care and end-of-life management.
APA:
Yip, J., Grießhammer, S., Leutheuser, H., Richer, R., Lu, H., Kölpin, A.,... Steigleder, T. (2026). Dead or Alive? Radar-Based Monitoring and Machine Learning to Identify Patient's Vital Status. (Unpublished, Submitted).
MLA:
Yip, Julia, et al. Dead or Alive? Radar-Based Monitoring and Machine Learning to Identify Patient's Vital Status. Unpublished, Submitted. 2026.
BibTeX: Download