Abstract
Abstract: :
Purpose: The goal of this project is to develop an educational resource that utilizes visual pattern recognition as a learning tool to expose training Ophthalmologists to the full gamut of eye diseases. Textbooks often provide single, ideal examples of disorders that may have great variability in appearance in the clinical setting. Moreover, many residents fail to make diagnoses due to a lack of exposure to certain diseases. We hypothesize that a comprehensive collection of images containing multiple examples of eye diseases can be created in the form of an electronic atlas and may greatly enhance diagnostic abilities. Methods: Thousands of unpublished clinical images were collected from specialists in the areas of Retina, Cornea, Oculoplastics and Glaucoma. These photos were scanned using a Nikon CoolScan 4000ED slide scanner into medium and high-resolution JPEG format. The resulting database containing images, findings, clinical information, and diagnoses was translated into static HTML format by a custom, automated software package. A Web server was created to make these HTML pages available on the Internet. Results: RedAtlas.org (Recognizing Eye Disease) is an online database that contains thousands of images of ophthalmic disorders, categorized by subspecialty (e.g. retina), disease category (e.g. white dot syndromes), disease entity (e.g. Best's disease), and clinical findings (e.g. intraretinal hemorrhage). The atlas spans the range of disorders seen in the field, including rare disorders that may not be seen in the normal course of residency training. Some images are organized into case presentations with clinical data while others stand alone within general categories for rapid viewing. Diseases are represented by multiple images and little accompanying text, emphasizing the importance of visual pattern recognition in making diagnoses in this field. The user is encouraged to reference standard textbooks for discussions of diseases. The web page format allows for easy retrieval of specific images or general browsing among categories. The atlas is easily updated and expandable because of its electronic format. Its content can be readily accessed for use in lectures and conferences, and it is freely accessible 24 hours per day around the world. Conclusion: A comprehensive atlas of clinical images can be an effective teaching resource in a field that relies highly on visual pattern recognition in making diagnoses. An Internet based electronic atlas is free and easily accessible to Ophthalmologists-in-training around the world.
Keywords: 430 imaging/image analysis: clinical • 455 learning • 432 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)