APPLICATIONS OF DEEP LEARNING TO IMPROVE THE QUALITY OF HEALTHCARE OUTCOMES

The aim of this paper is only to provide a systematic review of important research undertaken thus far in Deep Learning (DL) applications in healthcare and biomedicine. A total of 47 papers were shortlisted for this review. The review revealed that DL can be and has been applied in a number of healthcare contexts to improve outcomes. The two most important ones among them, indicated by the number of available papers, are cancer and medical imaging. A good number of papers on drugs and their development were also available. However, it must be highlighted that such categorisations may be somewhat arbitrary. It may be possible to categorise one paper in more than one category. The primary implication of this research for the academia is that there is a large deficit of papers on many of the chronic and lifestyle related illnesses such as diabetes as well as some diseases caused by immunodeficiency. There is also a complete deficit of literature on most of the acute health problems. This may be indicative of the need for more intensified research in the deficit areas. The primary implication of this research for health practitioners is that there is a plethora of substantial research that is currently available and accessible regarding the applications of DL to cancer and medical imaging which may be utilised for their practice.