Using Deep Learning for Medical Imaging with focus on Diabetic Retinopathy
With Machine Learning and Deep Learning bringing positive impact across various industries, it's progressively integrating into the Healthcare industry. One such notable case is detection of Diabetic Retinopathy in the field of Ophthalmology.
As many are now aware of Googles working with Eye hospital in India, we have found through one the top Retina specialist in south India with whom we are working closely that distinguishing between DR and No DR is the major primary need among doctors. We have been working closely in line with the inputs of our specialist doctor and testing out our algorithm in real time.
In this talk, I will be explaining the intricacies in capturing the patterns in Diabetic Retinopathy and how it has been helping the doctors to give dedicated time in handling patients effectively. We will further be touching on the Tensorflow framework which was used to build the model and choosing the right hardware.
Outline/Structure of the Case Study
-> Overview of Diabetic Retinopathy
- Cause of disease
- Stages of disease
- Occurrence Patterns
-> Technical Overview
- Extraction of disease patterns using computer vision
- Convnet architecture
- GPU’s used to compute this process
-> Benefits for Doctors & ROI
-> Extension of this process to Breast Cancer detection
- Understanding of how deep learning is applied in Medical Imaging.
- Learn how to build a Convolutional Neural Network.
- Understanding the challenges facing in Image data of Diabetic Retinopathy
Data Scientist, Machine Learning, Researchers and Healthcare professionals.
Prerequisites for Attendees
Understanding of Neural Networks and Image Processing.