The monthly PGRs Research Presentations was held on Wed. 8th January, 2pm, Room MC3108.
This session we had the following presentations:
| Title: “Intelligent Automated System for Diabetic Retinopathy Screening“. | Title: “Crowdsourcing through Social Media: Exploring and understanding crowdsourcing techniques in social media platforms” | |
|
By: Talal Albacha |
By: Obinna Ajuruchi |
|
| Abstract:Diabetic retinopathy (DR) is a common complication of diabetes. It damages the cells at the back of retina and if it is not treated, it can lead to blindness. Diabetes patients should have their eyes examined once a year for signs of damage, this periodic examination helps in early detection of the disease and protect the patient from blindness. The number of diabetes people in UK is 3,044,681 which equals around 4.6 % of UK population. This number highlights the huge cost of sponsoring the experienced ophthalmologists who can adequately grade DR of all patients across all cities in a timely manner.
Thus creating an automated system of detection and grading of DR can significantly reduce the cost and help in providing consistent service for all patients. This trend had been covered by multiple researches worldwide but still needs considerable enhancements and development especially in increasing specificity level and combination of multiple detectors in full automated and learning capable system. The aim of this research project is to build an intelligent system using image processing and machine learning techniques built-up within multi-phase analysis and scanning of the retinal image to help in increasing the accuracy level of the automated diabetic retinopathy detection and grading system; then to become a practical clinical experience |
Abstract: Crowdsourcing is the process of using large groups of unrelated people to solve a task or to analyse large amounts of data, which would otherwise take one person many hours to complete. Crowdsourcing platforms such as Amazon Mechanical Turk aims to use large numbers of people to achieve the same result in a much shorter time frame. Seti@home, whilst not using people per se, is an example of successful collaborative data analysis projects, with free computer CPU cycles being used to solve a task. Users of social network services have increased in recent years, as of March 2012 Facebook estimates its users at 900 million; These users are already online and are potential crowd workers. The objective of this research study is to investigate how users of social networks can be used in crowdsourcing scenarios, their motivations for doing so and the viability of existing methods. | |