Future Directions of Artificial Intelligence and Machine Learning in Healthcare: A Systematic Analysis and Mapping Study

In order to organise previously conducted research on using machine learning in medical contexts, the authors of this paper employ a method known as systematic mapping. In order to do this, we looked through the abstracts of a variety of publications, including medical journals, healthcare periodicals, and conference proceedings, looking for the phrase “use of artificial intelligence and machine learning in healthcare.” After doing a search on Google Scholar, which resulted in the retrieval of 500 papers, we classified these studies in accordance with their objectives, approaches, primary concerns, and illnesses. With the use of this technique, we were able to organise our findings into the five categories that are as follows: privacy and security; a framework for privacy and security; interpretable machine learning; medical image assessment; electronic health record processing; and transfer learning. In addition, we found that the evaluation of medical images is the topic that receives the most research, that interpretable machine learning and explainable artificial intelligence are becoming increasingly popular, and that most authors are primarily interested in cancer research. To restate our mission, we want to provide future generations of scholars with an accurate picture of where the field is now and where it is headed.