Potentials of Artificial Intelligence in Assessing Pancreatic Pathology Based on Spiral Computed Tomography Findings

Abstract

Artificial intelligence is the study of algorithms that give machines the ability to "reason" and acquire cognitive functions to achieve human–level performance in cognition-related tasks such as, for example, problem solving, object and word recognition, and decision-making. Currently, there are a lot of studies proving that artificial intelligence can not only diagnose diseases on a par with doctors, but also spend much less time on it. Artificial intelligence has entered many areas of medicine, and recently its role has become more significant in the diagnosis and treatment of pancreatic pathology. Over the past decade, the number and variation of methods for analyzing medical images has increased significantly due to the development of artificial intelligence, new programs for analyzing and systematizing objects. The aim of this review is to analyze, summarize and evaluate data published in the scientific literature on the use of artificial intelligence techniques to diagnose pancreatic pathology based on the results of computed tomography. It is demonstrated further perspectives and the need to develop this area in medical practice. A systematic literature search was conducted on the databases of the journals PubMed and eLibrary. The search for literature was carried out by Keywords"artificial intelligence", "pancreas", "computed tomography", "radiomics". The search interval was 2015-2023. The authors investigated all research studies of foreign and Russian scientists, which contain information on the use of diverse options of artificial intelligence techniques for differential diagnosis of pancreatic pathology, mainly based on computed tomography, and their assessment to demonstrate their further beneficial development in the field of medicine. To date, artificial intelligence programs based on spiral computed tomography data allow differentiating the pathology of the pancreas with high accuracy, which greatly facilitates human efforts and allows applying them as an indispensable assistant in work. That is why it is necessary to introduce these technologies into the circulation of medical institutions as actively as possible in order to expand the database of artificial intelligence, which will achieve more accurate results in the diagnosis of pancreatic diseases and more.