Artificial Intelligence and Medical Education

Abstract

Artificial Intelligence (AI) is a branch of computer sciences that uses learning algorithms to calculate probability of outcome by using Bayes theorem and other statistical methods for a given certain input (Fig.1). When the chance of an event occurring is calculated over and over again after adding new data or evidence at each step, the probability can reach the level of near certainty for given inputs. Thousands, even millions of data points are incorporated in calculating posterior probability for predictive analytics. The analytics are input neutral as programs predict the future events irrespective of the type of the data. AI has, thus, blurred the boundaries between the physical, digital, and biological worlds. The initial learning process is considered training where inputs are given to the program already marked for the expected outcome. This training information can either be highly precise or very vague allowing different degrees of freedom to the program but also increasing the burden of training. Once trained an AI algorithm is able to predict or analyze given input to suggest the required outcome with some certainty. This improves with continued training through feedback.