A general extensible learning approach for multi-disease recommendations in a telehealth environment
Article
Article Title | A general extensible learning approach for multi-disease recommendations in a telehealth environment |
---|---|
ERA Journal ID | 18106 |
Article Category | Article |
Authors | Lafta, Raid (Author), Zhang, Ji (Author), Tao, Xiaohui (Author), Zhu, Xiaodong (Author), Li, Hongzhou (Author), Chang, Liang (Author) and Deo, Ravinesh (Author) |
Journal Title | Pattern Recognition Letters |
Journal Citation | 132, pp. 106-114 |
Number of Pages | 9 |
Year | 2020 |
Publisher | Elsevier |
Place of Publication | Netherlands |
ISSN | 0167-8655 |
1872-7344 | |
Digital Object Identifier (DOI) | https://doi.org/10.1016/j.patrec.2018.11.006 |
Web Address (URL) | https://www.sciencedirect.com/science/article/abs/pii/S016786551830881X |
Abstract | In a telehealth environment, intelligent technologies are rapidly evolving toward improving the quality of patients’ lives and providing better clinical decision-making especially those who suffer from chronic diseases and require continuous monitoring and chronic-related medical measurements. A short-term disease risk prediction is a challenging task but is a great importance for teleheath care systems to provide accurate and reliable recommendations to patients. In this work, a general extensible learning approach for multi-disease recommendations is proposed to provide accurate recommendations for patients with chronic diseases in a telehealth environment. This approach generates appropriate recommendations for patients suffering from chronic diseases such as heart failure and diabetes about the need to take a medical test or not on the coming day based on the analysis of their medical data. The statistical features extracted from the sub-bands obtained after a four-level decomposition of the patient's time series data are classified using a machine learning ensemble model. A combination of three classifiers – Least Squares-Support Vector Machine, Artificial Neural Network, and Naive Bayes – are utilized to construct the bagging-based ensemble model used to produce the final recommendations for patients. Two real-life datasets collected from chronic heart and diabetes disease patients are used for experimentations and evaluation. The experimental results show that the proposed approach yields a very good recommendation accuracy and offers an effective way to reduce the risk of incorrect recommendations as well as reduces the workload for chronic diseases patients who undergo body tests most days. Thus, the proposed approach is considered one of a promising tool for analyzing time series medical data of multi diseases and providing accurate and reliable recommendations to patients suffering from different types of chronic diseases. |
Keywords | chronicle heart disease; dual-tree complex wavelet transformation; machine learning ensemble; recommender system; telehealth environment; time series prediction |
ANZSRC Field of Research 2020 | 469999. Other information and computing sciences not elsewhere classified |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Byline Affiliations | School of Agricultural, Computational and Environmental Sciences |
Nanjing University of Information Science and Technology, China | |
Guilin University of Electronic Technology, China | |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q59yx/a-general-extensible-learning-approach-for-multi-disease-recommendations-in-a-telehealth-environment
231
total views10
total downloads14
views this month0
downloads this month