A machine learning approach for mortality prediction only using non-invasive parameters
- PDF / 2,457,984 Bytes
- 44 Pages / 595.276 x 790.866 pts Page_size
- 18 Downloads / 184 Views
REVIEW ARTICLE
A machine learning approach for mortality prediction only using non-invasive parameters Guang Zhang 1 & JiaMeng Xu 1 & Ming Yu 1 & Jing Yuan 1 & Feng Chen 1 Received: 23 October 2019 / Accepted: 26 March 2020 # International Federation for Medical and Biological Engineering 2020
Abstract At present, the traditional scoring methods generally utilize laboratory measurements to predict mortality. It results in difficulties of early mortality prediction in the rural areas lack of professional laboratorians and medical laboratory equipment. To improve the efficiency, accuracy, and applicability of mortality prediction in the remote areas, a novel mortality prediction method based on machine learning algorithms is proposed, which only uses non-invasive parameters readily available from ordinary monitors and manual measurement. A new feature selection method based on the Bayes error rate is developed to select valuable features. Based on non-invasive parameters, four machine learning models were trained for early mortality prediction. The subjects contained in this study suffered from general critical diseases including but not limited to cancer, bone fracture, and diarrhea. Comparison tests among five traditional scoring methods and these four machine learning models with and without laboratory measurement variables are performed. Only using the non-invasive parameters, the LightGBM algorithms have an excellent performance with the largest accuracy of 0.797 and AUC of 0.879. There is no apparent difference between the mortality prediction performance with and without laboratory measurement variables for the four machine learning methods. After reducing the number of feature variables to no more than 50, the machine learning models still outperform the traditional scoring systems, with AUC higher than 0.83. The machine learning approaches only using non-invasive parameters achieved an excellent mortality prediction performance and can equal those using extra laboratory measurements, which can be applied in rural areas and remote battlefield for mortality risk evaluation. Keywords Early mortality prediction . Machine learning . Non-invasive parameters . Feature reduction
1 Introduction Mining and analyzing multiple physiological parameters of patients and predicting their mortality can provide a timely and necessary medical intervention, to rationally dispose medical resources and decrease mortality. There are many traditional scoring systems in clinical to evaluate the intensive care patient’s condition and mortality, such as the simplified acute physiology score (SAPSII) [1], sequential organ failure assessment (SOFA) [2], and acute physiological score (APS) [3]. The traditional scoring system * Guang Zhang [email protected] * Feng Chen [email protected] 1
Institute of Medical Support, Academy of Military Sciences, Tianjin, China
performs weighted methods based on the demographic parameters; non-invasive physiological parameters; and laboratory measurements reflecting the patient’s dise
Data Loading...
