Artificial Intelligence in endoscopy: A future poll

Before 1950, Artificial Intelligence [AI] was considered as a glimpse of the future. Now, it is considered a mandated technology that is included in our everyday life [1]. AI can be defined as any technique that enables computers to mimic human cognitive functions using logical reasoning, if-then rules, decision trees, and deep machine learning [2]. It is crucial to highlight basic definitions as machine learning [ML], which is a subset of AI that abstruse statistical techniques that allow improving machine performance with experience [3]. While deep learning [DL] is a subset of machine learning consists of algorithms that enable the software to train itself to improve the performance of their tasks as in image and speech recognition. DL can be achieved by exposing the algorithms' multilayered networks to a considerable amount of data [4]. Such AI techniques are used for what is called “Big Data Analytics,” where raw data is vastly unlabeled and uncategorized [5]. Since 2010, substantial progress has been made in the AI field, which unlocked many solutions for intensive medical problems using big data analytics [3]. In endoscopy, the doctor has to navigate through the tissue to identify, diagnose visually, and screen for suspicious lesions and diseases. This is a complex intensive task that requires a myriad of clinical skills, especially in the critical, unstable condition, which would lead to missing target lesions [6]. With the advancement of imaging techniques, doctors are facing a serious challenge in processing high volume full visual data, which could reach 30 high definition frames per second during real-time endoscopy [7]. AI would help better visualize and characterize lesions and assist doctors in better clinical decision making [8]. The aim of the work is to standardize AI terminologies and introduce definitions in the wake of the expansion of AI applications in the field of endoscopy. Also, we aim to highlight to which extent this technology has succeeded in endoscopy, how can we benefit from it and what are the future directions.

留言 (0)

沒有登入
gif