USER INTERFACE DESIGN OF WEB-BASED APPLICATION FOR TRAINING ON ABNORMALITIES DETECTION OF FUNDUS IMAGES USING MANUAL ANNOTATION METHOD

• Main Author: Akhsanil Fitri, Navila
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/42604
• Institution: Institut Teknologi Bandung

More than 10 million Indonesians who are affected by Diabetes, until the prevalence rate in Indonesia reaches 6.2%. Besides, in 2013 Indonesia was categorized as one of the countries with the highest number of people with Diabetes. One complication due to this type of diabetes is the occurrence of a visual disorder called Diabetic Retinopathy (DR). So it takes a lot of competent ophthalmologists to continue monitoring to prevent worse complications of DR. Therefore, resident doctors in ophthalmology need a learning tools to improve their ability to make a diagnosis in quick and accurate way. One solution for practicing diagnosis is to utilize digital fundus images. A web application called Optan has been developed as a tools to diagnose fundal image abnormalities using manual annotation by users. The functions used on the Optan web annotation adapt the code from a web-based annotation tools named VIA, that developed by the Visual Geometry Group, University of Oxford. The web development component of VIA is made using HTML5, CSS, and Javascript. The Optan web application interface is designed with a simple layout so that it is easily understood by users. A front-end framework Bootstrap mostly used for the Optan’s development. To assess usability of the application, a survey to 16 respondents from three different professional backgrounds was conducted. Respondents were asked to try the Optan application and fill out the QUIS questionnaire to assess satisfaction of the interface of the system. The survey show that the user interface of the Optan web application is quite easy to be understoodd and used with an average score of 7.3 from the range 0 to 9 grade.