Deep Learning Paves Way for Better Diagnostics

By Tiffany Trader

September 19, 2016

Stanford researchers are leveraging GPU-based machines in the Amazon EC2 cloud to run deep learning workloads with the goal of improving diagnostics for a chronic eye disease, called diabetic retinopathy. The disease is a complication of diabetes that can lead to blindness if blood sugar is poorly controlled. It affects about 45 percent of diabetics and 100 million people worldwide, many in developing nations.

Final-year Stanford PhD students Apaar Sadhwani and Jason Su got involved in developing the diagnostic solution as part of a class project and corresponding Kaggle competition that was held last year. Sponsor Amazon provided AWS cloud credits in support of the research.

diabetic-retinopathy_5-classes_sadhwani-su_400x
Source: Automatic Grading of Eye Diseases Through Deep Learning, 2016

After Kaggle, the duo decided to turn their research project into a cloud-based platform that hospitals and clinics can use to guide the diagnosis of eye diseases. Their approach relies on a convolutional neural net (CNN) that grades the severity of diabetic retinopathy disease states into five categories: 0-4, with 0 being normal and 4 being the most severe.

The researchers have been training their model with a data set of 80,000 images from EYEPACS, a web-based application for exchanging eye-related clinical information, run by the California Health Foundation. “Getting data is the most constraining part of applying deep learning to a medical setting,” said Sadhwani, “but we are working closely with partners to get more data.”

They’ve also had to address a class imbalance in the data set. “We have a lot more 0’s and 1’s than 3’s and 4’s, for example,” said Sadhwani. As the disease progresses to stage four (known as proliferative diabetic retinopathy, or PDR), image data is more rare. A total of about 10,000 stage four images are required for optimal results.

The training problem is run on AWS Elastic Compute Cloud (EC2) with single-GPU and multi-GPU nodes. Some S3 storage and Elastic Block Store (EBS) services are also employed. The training takes about three days to a week for a given model.

Within EC2, the researchers are using Starcluster which lets them build custom clusters among the nodes and network them together. They used a master node to store all their training data and up to 28 different training nodes. All these separate training nodes would access the master node so they wouldn’t have to mirror the data onto each of the nodes.

“With Starcluster and AWS you can bring up different node types independently on demand,” said Su. “So we would run this experiment that would only need a single-GPU node and then after that finished we could shut down that node and save money. Then we would scale it up to a larger resolution image and we would need four-GPU nodes for that – so we’d spin that up, train on that, and come back three days later and shut that off. AWS provides this flexibility for scaling up and scaling down for cost and for trying out different ideas.”

The researchers relied on AWS spot instance pricing to further improve the economics. Their program saves a state every “epoch,” which relates to one pass through the data set, so losing a node did not incur a big setback. With 55 epochs in a run, the most they would lose is 1/55th of their training progress.

They used the g2.2xlarge instance type and the g2.8xlarge instance type for training their final models. They trained two kinds of models, one on low-res images and the final model on high-res images, for which they employed the larger multi-GPU nodes.

Amazon’s GPU instances are based on older Nvidia GRID K520 graphics cards, which at 4 GB per GPU do not have an ideal memory profile for training based on very high-resolution images.

“Typically in deep learning, you have a 256×256 image, or about one-sixteenth of a megapixel and we’re at four megapixels, so memory is a huge part of doing this problem,” said Sadhwani. “Our workaround was to scale to 4-GPU nodes, which effectively had 4 gigabytes of memory each [GPU], but we lose some to overhead because we have to have the model independently at each of the separate GPUs. It would be more advantageous to have a single GPU with a full 16 gigabytes.”

Because their model was dealing with these high-resolution images, they used Torch to split it across the 4-GPU node to fine-tune its parameters. Currently, they are moving to a distributed training model, which enables several different nodes to train essentially the same model but with independent data. This gives them the ability to train one model across many GPUs, rather than a single model on a single GPU node and thus accelerates the training.

The researchers are eyeing clouds with higher-memory GPUs, which could mean holding out for upgraded Amazon instances or moving to the Microsoft Azure cloud with its Tesla K80s.

They are not interested in CPUs. “It would take significantly longer, at least a factor of 50,” said Sadhwani. “The kind of neural networks we are using [convolutional neural nets] harness parallelization a lot. Even if we were not using this special class of network, there is at least a 10x speedup going from CPUs to GPUs, but for this particular variety that speedup is magnified a lot more, in the neighborhood of 100x.”

Diabetic retinopathy is a disease of the blood vessels in the eye. As the sugar level in the blood rises, it causes the walls of the blood vessels to thin and eventually they’ll crack and bleed. The most important thing to look for is tiny dot bleeds, called hemorrhages. They are very small and difficult to locate even with advanced algorithms. The deep learning model must also be trained to ignore or flag likely camera artifacts, which appear in approximately 40 percent of the images, and can obscure identification of disease traits.

To address these challenges, the Stanford team’s approach uses two networks, a lesion detector and a main network. The lesion detector looks at a small part of the image and outputs a number between 0 and 1, a probability. The lesion detector has so far achieved an accuracy of 99 percent for negatives and 76 percent for positives. The purpose of the main network is to characterize details about where the disease-related features are with respect to the important parts of the eye.

deep-learning-fused-architecture_sadhwani-su_800x
Source: Automatic Grading of Eye Diseases Through Deep Learning, 2016

The outputs of these two pipelines are then fused together. This provides a way to combine low-level details about where there are dot hemorrhages with high-level information like which parts of the image should actually be ignored because they are corrupted by artifacts. The fuse network is responsible for integrating all these signals together to deliver a final probability for the disease class.

Right now the team has been working with five classes, but they say that in the clinical setting, these grades are not tracked with such granularity. In terms of intervention, there are really three stages: 0) no action is required; 1) monitor the progress of the disease; and 2) medical intervention such as surgery is required.

“Moving to three-classes would increase the accuracy of our models because it’s a simpler problem and easier to solve,” said Su.

The ultimate goal here is to deliver a digital assistant to radiologists, opthamologists and other clinicians, so they can screen more patients, more frequently.

“Using an automated tool to augment human resources, you can more closely monitor the changes in the disease state as they progress to more effectively treat the disease,” said Su.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understanding on January 10. The MOU represents the continuation of a 1 Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Tennessee), Satoshi Matsuoka (Tokyo Institute of Technology), Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown and Spectre security updates on the performance of popular H Read more…

By Tiffany Trader

HPE Extreme Performance Solutions

HPE and NREL Take Steps to Create a Sustainable, Energy-Efficient Data Center with an H2 Fuel Cell

As enterprises attempt to manage rising volumes of data, unplanned data center outages are becoming more common and more expensive. As the cost of downtime rises, enterprises lose out on productivity and valuable competitive advantage without access to their critical data. Read more…

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension around the potential changes that could affect or disrupt Lustre Read more…

By Carlos Aoki Thomaz

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understandi Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Te Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown Read more…

By Tiffany Trader

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension aroun Read more…

By Carlos Aoki Thomaz

When the Chips Are Down

January 11, 2018

In the last article, "The High Stakes Semiconductor Game that Drives HPC Diversity," I alluded to the challenges facing the semiconductor industry and how that may impact the evolution of HPC systems over the next few years. I thought I’d lift the covers a little and look at some of the commercial challenges that impact the component technology we use in HPC. Read more…

By Dairsie Latimer

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

Momentum Builds for US Exascale

January 9, 2018

2018 looks to be a great year for the U.S. exascale program. The last several months of 2017 revealed a number of important developments that help put the U.S. Read more…

By Alex R. Larzelere

ANL’s Rick Stevens on CANDLE, ARM, Quantum, and More

January 8, 2018

Late last year HPCwire caught up with Rick Stevens, associate laboratory director for computing, environment and life Sciences at Argonne National Laboratory, f Read more…

By John Russell

Inventor Claims to Have Solved Floating Point Error Problem

January 17, 2018

"The decades-old floating point error problem has been solved," proclaims a press release from inventor Alan Jorgensen. The computer scientist has filed for and Read more…

By Tiffany Trader

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Japan Unveils Quantum Neural Network

November 22, 2017

The U.S. and China are leading the race toward productive quantum computing, but it's early enough that ultimate leadership is still something of an open questi Read more…

By Tiffany Trader

AMD Showcases Growing Portfolio of EPYC and Radeon-based Systems at SC17

November 13, 2017

AMD’s charge back into HPC and the datacenter is on full display at SC17. Having launched the EPYC processor line in June along with its MI25 GPU the focus he Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

IBM Begins Power9 Rollout with Backing from DOE, Google

December 6, 2017

After over a year of buildup, IBM is unveiling its first Power9 system based on the same architecture as the Department of Energy CORAL supercomputers, Summit a Read more…

By Tiffany Trader

Fast Forward: Five HPC Predictions for 2018

December 21, 2017

What’s on your list of high (and low) lights for 2017? Volta 100’s arrival on the heels of the P100? Appearance, albeit late in the year, of IBM’s Power9? Read more…

By John Russell

Chip Flaws ‘Meltdown’ and ‘Spectre’ Loom Large

January 4, 2018

The HPC and wider tech community have been abuzz this week over the discovery of critical design flaws that impact virtually all contemporary microprocessors. T Read more…

By Tiffany Trader

Leading Solution Providers

Perspective: What Really Happened at SC17?

November 22, 2017

SC is over. Now comes the myriad of follow-ups. Inboxes are filled with templated emails from vendors and other exhibitors hoping to win a place in the post-SC thinking of booth visitors. Attendees of tutorials, workshops and other technical sessions will be inundated with requests for feedback. Read more…

By Andrew Jones

Tensors Come of Age: Why the AI Revolution Will Help HPC

November 13, 2017

Thirty years ago, parallel computing was coming of age. A bitter battle began between stalwart vector computing supporters and advocates of various approaches to parallel computing. IBM skeptic Alan Karp, reacting to announcements of nCUBE’s 1024-microprocessor system and Thinking Machines’ 65,536-element array, made a public $100 wager that no one could get a parallel speedup of over 200 on real HPC workloads. Read more…

By John Gustafson & Lenore Mullin

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Flipping the Flops and Reading the Top500 Tea Leaves

November 13, 2017

The 50th edition of the Top500 list, the biannual publication of the world’s fastest supercomputers based on public Linpack benchmarking results, was released Read more…

By Tiffany Trader

GlobalFoundries, Ayar Labs Team Up to Commercialize Optical I/O

December 4, 2017

GlobalFoundries (GF) and Ayar Labs, a startup focused on using light, instead of electricity, to transfer data between chips, today announced they've entered in Read more…

By Tiffany Trader

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Nvidia, Partners Announce Several V100 Servers

September 27, 2017

Here come the Volta 100-based servers. Nvidia today announced an impressive line-up of servers from major partners – Dell EMC, Hewlett Packard Enterprise, IBM Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This