Affiliations: [a] Department of Biomedical Engineering, University of Miami, Coral Gables, FL, USA | [b] Division of Gastroenterology, University of Miami, Leonard Miller School of Medicine, Miami, FL, USA | [c] Department of Medicine, Phramongkutklao College of Medicine, Bangkok, Thailand | [d] Department of Electrical and Computer Engineering, University of Miami, Coral Gasbles, FL, USA
Correspondence:
[*]
Address for correspondence: Weizhao Zhao, PhD, Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA. Tel.: +1 3052846763; Fax: +1 3052842404; E-mail: [email protected]/[email protected].
Abstract: Competence in performing endoscopy requires a considerable amount of hands-on practice for a physician to master. In an effort to reduce the involvement of patients in the training process, simulation has become an important part of endoscopic training. This paper presents a low-cost training and evaluation system for upper endoscopy that serves as an additional or alternative tool, particularly for entry-level gastroenterology fellows. The protocol of making a customized mechanical training model that provides realistic visual appearances of the upper GI tract was created. The materials and methods used in the fabrication are outlined step-by-step in this report. A software application associated with the mechanical model was also developed to provide help for trainee during practice, evaluate the performance automatically upon completion of the training tasks, and record the performance and results in a database for review or further analysis by the trainee or instructor. Our software provides trainee with an interactive training experience. A survey obtained from sixteen gastroenterologists shows a promising feasibility of using the developed system. The developed low-cost upper endoscopy simulator can provide both training and evaluation functions with substantially low investment.
Keywords: Upper endoscopy, training simulator, fellow evaluation, endoscopic image classification, support vector machines
