The Application of Artificial Intelligence Technology in Healthcare: A Systematic Review

Abstract

© 2020, Springer Nature Switzerland AG. The proliferation of artificial intelligence and its continued development can be attributed to the pursuit of advanced machine learning techniques for handling big health data. Even though AI appears to be an independent system while considering algorithms and learning techniques, it, however, requires integration of different machine learning algorithms to enable it to handle different data structures. Notably, the number of articles addressing AI implementation from a medical research perspective are on the rise. Further, AI in medical research have machine learning component and as such relies on algorithms such as support vector machine, neural network, deep learning, and convolution neural networks. Of these algorithms, support vector machine is the most commonly used, and it has been applied in medical imaging, diagnosis and treatment of stroke as well as early detection of cancer and neurology conditions. As per the survey, AI results in higher accuracy of diagnosis and risk prediction compared to human approaches. Despite such success and promising future, AI faces regulatory and data related challenges.