ARTIFICIAL INTELLIGENCE AND ADVANCES IN DIAGNOSTIC IMAGING IN RADIOLOGY

Abstract

Introduction: Artificial intelligence (AI) has been emerging as a complement to the growing role of radiology in diagnostic and interventional medicine, reducing stress points for radiology professionals. Objective: To identify in the available literature the main advances that artificial intelligence has been providing in diagnostic imaging in radiology. Methodology: This is a systematic review, whose searches were performed in the virtual libraries MEDLINE via Pubmed and Web of Science, and in the LILACS and CINAHL databases. Cross-reference searches were also performed. For the search, controlled descriptors and the Boolean operators AND, for simultaneous occurrence of subjects, and OR, for occurrence of one or another subject, were adopted. The descriptors used were: "Artificial Intelligence", "Radiology" and "Diagnostic Imaging". To better organize the sample collection, we opted for the use of advanced search. The descriptors were combined with each other with the Boolean connector OR and AND, within each set of terms of the PICo strategy, and then crossed with the Boolean connector. Results: It is highlighted that AI allows the possibility of combining several sources of information, in addition to the image, to obtain a more accurate diagnosis, helping to increase productivity and better management in industry dynamics and accurate diagnosis. Final considerations: It is concluded that the knowledge and practices acquired by professionals of radiological techniques in performing minimally invasive exams and procedures, are a basic and necessary factor for the radiology sector where the AI will become a great ally of health professionals.