The advent of eHealth and the need for real-time patient monitoring and assessment has prompted interest in understanding people behavior for improving care services. In this paper, the application of machine learning algorithms in clustering and predicting vital signs was pursued. In the context of big data and the debate surrounding vital signs data is fast becoming more relevant and applicable in predictive medicine. This paper assesses the applicability of k-Means and x-Means in clustering signals and used deep learning, Naïve Bayes, Random Forests, Decision Trees, and Generalized Linear Models to predict human dynamic motion-based vital signal patterns.
|Title of host publication||Studies in Computational Intelligence|
|Number of pages||24|
|Publication status||Published - 4 Sep 2019|
|Name||Studies in Computational Intelligence|
Bibliographical noteFunding Information:
We are grateful to the UCI team for granting access to the data used in the study. We acknowledge and appreciate the Oresti Banos, Rafael Garcia, and Alejandro Saez of the Department of Computer Architecture and Computer Technology, University of Granada for collecting and sharing the data with UCI.
© Springer Nature Switzerland AG 2020.
- Cognitive hypervisor
- Deep learning
- Machine learning
- Vital signals