Diabetic Retinopathy Detection

DEADLINE: July 20, 2015

Diabetic retinopathy is a real-world problem that can benefit from a simulation-based solution. Join the Kaggle competition to develop an automated detection program. Entries are due July 20.

Many diabetics in California do not get screened for diabetic retinopathy, a sight-threatening complication of the disease, due to high cost and limited access. In 2007 CHCF launched a project that provided special cameras to safety-net clinics to capture and electronically transmit retinal images to optometrists and ophthalmologists who would screen the images from afar. While this project dramatically improved access to screening, it did not reduce the costs nor the time-consuming, manual process of screening an image.

Researchers are developing computer algorithms that could screen a retinal image for diabetic retinopathy, resulting in faster, more cost effective, and potentially more accurate readings. To catalyze advancement in this field, CHCF has partnered with Kaggle, a platform for predictive modeling and analytics competitions on which sponsors post their data, and computer scientists, statisticians, engineers, and data miners from all over the world compete to produce the best models.

The Diabetic Retinopathy Detection competition will award $100,000 among the top three teams that use image classification, pattern recognition, or machine learning to develop an automated diabetic retinopathy detection model with realistic clinical potential. Submissions are due July 20, 2015.

Competitors will have access to a large set of high-resolution retinal images from EyePACS, a free platform for retinopathy screening, that are scored for the presence of diabetic retinopathy on a five-point scale.

Submissions will be evaluated by the algorithm's ability to accurately identify and score the level of retinopathy. For complete rules, see the Kaggle competition website.

RELATED CHCF PAGES

DiabeticRetinopathy Screening

EXTERNAL LINKS

KaggleCompetition: Diabetic Retinopathy Detection

Source: California Healthcare Foundation

Share to Other !