How Boston Children’s Hospital Augments Doctors Cognition with Red Hat OpenShift

Software can be an enabler for healers. At Red Hat, we’ve seen this first hand from customers like Boston Children’s Hospital. That venerable infirmary is using Red Hat OpenShift and Linux containers to enhance their medical capabilities, and to augment their doctors cognitive capacity.

The Mother of all Demos

Let’s rewind 50 years to Silicon Valley as a quiet suburb with a big college doing very strange things for its day. We’re now back at Stanford Research Institute, also known as SRI. This facility, in December of 1968, hosted the Mother of all Demos: A revolutionary and ground breaking technology demo.

The Mother of all Demos was run by Douglas Engelbart, the head of SRI’s Augmentation Research Center. Back in 1969, this group demonstrated the following things on stage for the very first time: the Mouse, the GUI, the word processor, the Internet, the collaborative Internet-based Word Processor, the on-screen cursor, the chording keyboard and telepresence. You could also make a compelling argument that SRI invented the computer monitor as the standard visual interface.


Engelbart envisioned a world where humans use computers to expand their consciousness and their data analysis capabilities. Computers were not seen by SRI as the means to an end, but rather as tools to allow humans to become even more curious, experimental and exploratory. They were tools to increase productivity and mental capacity, not just for doing your taxes or counting things.

Today, however, the breakthroughs are much more modest by comparison to that original Mother of all Demos. In a way, we’re only now realizing the second phase of Engelbart’s dream the first being the proliferation of desktop computing and word processing. Beyond those mundane compute tasks, however, we see Google as the force that has most augmented human cognition with its search engine, today.

That doesn’t mean the second phase of human cognition enhancement hasn’t taken place, however. It’s just not being written about as much as those flashy face swapping applications.

The Future is Now

When Dr. Ellen Grant, Director of an innovative research team at Boston Children’s Hospital took the stage at Red Hat Summit in Boston earlier this year, she was able to show the incredible power computing can unleash when it is coupled with human knowledge and human cooperation.

ChRis, AKA the Research Integration Service, was created by a collaboration between the Boston Children’s Hospital, Boston University, and Red Hat. The system uses the scalability and flexibility of Red Hat OpenShift to provide an infrastructure of storing, analyzing and sharing patient data between doctors across a number of hospitals.

Dr. Grant explained how the system helps her diagnose a patient. Without ChRis, if a new patient presenting unexplained seizures showed up in her office, she’d spend 15 to 20 minutes looking through around 5,000 images of the child’s brain generated by an MRI. After that, she’d forward the patient to other doctors for extensive tests and medication, none of which are guaranteed to fix the problem.

With ChRis, Dr. Grant would instead first process those 5,000 images at scale across the compute cluster. The resulting data would add information to those images, such as coloring the various regions of the brain, and highlighting areas where the brain structure has more than a standard deviation from the normal.

Dr. Grant can then compare these MRIs to those of other patients with similar symptoms, even though that data has been anonymized to protect the patients’ identities. The information on those patients is not confined to simply those who’ve attended Boston Children’s Hospital, either: many hospitals can share ChRis, and share their data with other facilities. They call these other hospital datacenters “Enclaves.”

This is tremendously important, said Dr. Grant, as children’s hospitals often see extremely rare diseases and afflictions, making data on those cases scarce. It’s almost incumbent upon these hospitals to share their patient information safely, as the sample pool for some medical issues is far below the needed threshold for proper statistical analysis.

Dr. Grant laid out a hypothetical scenario around this fictional seizure patient, and she intimated that without ChRis, the child could be in for a lifetime of seizures and ineffective medical treatments, all because the initial examination would be from a doctor who had only 15 to 20 minutes to find a defect in the child’s brain MRIs.

“Now this is the future of precision medicine: this is what we want to do and this is not possible without the Red Hat infrastructure and ChRis to bridge those two worlds together. Our lead engineer is working hand-in-hand with the Red Hat engineers so there’s no black boxes, and that’s another critical point in medicine: I need to know what happens to my data. I need to trace it through so that I understand the analysis that I get. Working together in open source, yet encrypted environments has now helped us share our collective knowledge to better serve and save while protecting individual identity. Together we are changing how healthcare works and it’s about time,” said Dr. Grant.

Red Hat is enabling the Boston Children’s Hospital to have the tools they need to save lives and innovate in medicine. That’s their business, after all. We’re here to provide the infrastructure to allow them to do their work in a better, more productive and more impactful manner. But the innovation they’ve unlocked has been thanks to their skilled technical and medical teams.

While SRI has since moved on to a much wider range or research than took place in Engelbart’s time, we see our work on open source, Linux, Kubernetes and OpenShift as an extension of that 50 year old dream. We’re here to help augment the capabilities of humans around the world, and in doing so, enable those humans to cure many medical diseases and save lives.

Learn more in the press release.

Categories
AI/ML, Containers, Events, OpenShift Ecosystem
Tags
, , , , ,