Text mining in radiology reports (Methodologies and algorithms), and how it affects on workflow and supports decision making in clinical practice (Systematic review)

The purpose of this review was to summarize the algorithms and methodologies of text-mining and demonstrate the main objective of text-mining on radiology reports in health care facilities which consider as a common source of medical information. And how it affects radiologist performance and plays a big role in clinical practice workflow and decisions making in critical situations and time-consuming. In case the radiologist widely used a narrative- text box in their reporting and sometimes there is a big need to know very specific and critical information about the patients’ current status and to provide with accurate diagnosis then take the appropriate action as soon as possible. However, here it becomes the need to utilize information technology and the effort was directed to find ways to merge data science with the health care field to solve such a problem. We follow the systematic review methodology conducted by Ahmad Alaiad .et al study completed after 29 quantitative and systematic related articles were searched using relevant database then extract and discuss the text-mining processes and provide overall picture about such new innovation and how the IT now days play a valuable role in health problem solving and make the clinical practice more effective and efficient and improve quality of care by improving clinical decision making process. We develop a research taxonomy that summarizes the most of algorithms and methodologies of existing research, we identify the major future questions, limitations and gaps.