Staging Cloud Platform for Deep Learning projects

Project Description Medical image processing with deep learning:

Our research involves developing an AI system for assisting radiologists with expeditiously diagnosing serious health concerns by scanning through images or other data. Although this type of research is in its early stages, it seems to offer great potential. Here are some sample research work and resources [1, 2, 3]
In the initial phase we use publicly available data to explore the:
- Feasibility
- Utility
- Effectiveness
with various anomaly detection machine learning algorithms, including various architectures of neural networks. A vast repository of data is publicly available on sites like https://nihcc.app.box.com/ (National Institutes of Health Clinical Center) and https://www.kaggle.com/c/diabetic-retinopathy-detection/data

The dataset involved is fairly large and deep learning is compute-intensive and we want to initiate the migration of our work to a scalable platform, the cloud, where resources are available on an as-needed basis. A qualified student will be able to complete the migration in 3 to 4 months of time.

References:
1. Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, MohammadhadiBagheri, Ronald M. Summers.ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases, IEEE CVPR, pp. 3462-3471,2017

2. https://nanonets.com/blog/deep-learning-for-medical-imaging/

3. https://radiology.ucsf.edu/blog/ai-rivals-expert-radiologists-detecting-brain-hemorrhages