Toward analyzing and synthesizing previous research in early prediction of cardiac arrest using machine learning based on a multi-layered integrative framework.

BACKGROUND: One of the significant problems in the field of healthcare is the low survival rate of people who have experienced sudden cardiac arrest. Early prediction of cardiac arrest can provide the time required for intervening and preventing its onset in order to reduce mortality. Traditional statistical methods have been used to predict cardiac arrest. They have often analyzed group-level differences using a limited number of variables. On the other hand, machine learning approach, which is part of a growing trend of predictive medical analysis, has provided personalized predictive analyses on more complex data and produced remarkable results.

OBJECTIVE: This paper has two aims. First, it offers a systematic review to evaluate the capability and performance of machine learning techniques in predicting the risk of cardiac arrest. Second, it offers an integrative framework to synthesize the researches in this field.

METHOD: A systematic review of cardiac arrest prediction studies was carried out through Pubmed, ScienceDirect, Google Scholar and SpringerLink databases. These studies used machine learning techniques and were conducted between the years 2000 and 2018.

RESULTS: From a total of 1617 papers retrieved from the literature search, 75 studies were included in the final analysis. In order to explore how machine learning techniques were employed to predict cardiac arrest, a multi-layered framework was proposed. Each layer of the framework represents a classification of the current literature and contains taxonomies of relevant observed information. The framework integrates these classifications and illustrates the relative influence of a layer on other layers. The included papers were analyzed and synthesized through this framework. The used machine learning techniques were evaluated in terms of application and efficiency. The results illustrated the prediction capability of machine learning methods in predicting cardiac arrest.

CONCLUSION: According to the results, machine learning techniques can improve the outcome of cardiac arrest prediction. However, future research should be carried out to evaluate the efficiency of rarely-used algorithms and to address the challenges of external validation, implementation and adoption of machine learning models in real clinical environments.