Page 44 - FCW, November/December 2021
P. 44

services of any network, such as DNS and [Dynamic Host Configuration Pro- tocol], but it also brings in modern technology that otherwise would not be available to offline computers, such as exploratory data analysis and visu- alization,” said Charles Avery, chief of the Program Services Branch in CDC’s Office of the CIO.
Each of CDC’s 200-plus labs has its own objectives, and it’s not unusual for half the lab equipment to be on the busi- ness network and the other half oper- ating offline. Avery said data is often transferred between computers via bio- metric-based thumb drives. ISLE moves everything onto the same autonomous network and applies security controls to allow for the movement of data.
“That simple change is huge for the lab scientists who are trying to con- stantly capture, analyze and review data, and you couldn’t do that with- out having the computers networked in some manner,” Avery said.
CDC has moved all the labs on its Chamblee, Ga., campus — including about 770 computers and other equip- ment — onto ISLE, and the feedback from lab workers has been favorable, said Wendi Kuhnert, senior advisor for laboratory science to the deputy direc- tor for infectious diseases at CDC.
“Especially with sequencing and where we are with our ability to col- lect data, we need the same capacity to analyze and move and share our data in a secure way,” Kuhnert said. “As data gets bigger and bigger, this has become much more important. That’s definitely a benefit we’re already seeing and will continue to see as we move forward.”
Officials are now connecting the labs at CDC’s Atlanta headquarters before moving on to the other labs.
Machine Learning-as- a-Service Platform Food and Drug Administration
The Food and Drug Administration has long relied on a manual, labor-intensive
process for inputting the large amount of handwritten documents and forms it receives. Those documents are often in languages other than English and arrive in many formats, including PDFs, Microsoft Word and Excel doc- uments, and photographs and scans, which may be blurry.
“One of the challenges that we faced was the amount of time that it takes to get some sort of structured data out of those forms and then really do something with that data in terms of our public health mission,” said Sohail Chaudhry, the FDA’s acting CTO.
To speed the process and increase the accuracy of data input, the FDA developed a machine-learning-as-a- service (MLaaS) platform. The cloud- delivered platform offers pre-built models, algorithms and robotic process automation tools, along with computer vision, image classification and natural language processing technology.
ML-based tools can recognize a form and its type, decipher handwrit- ten content and digitize it. They can also translate foreign languages into English, extract key information and apply it to a downstream application.
“The fact that it’s offered in our pre- authorized cloud, the solution itself is low in cost, [and] it increases our deploy- ment flexibility because most of our next- gen services and applications are being deployed in the cloud,” Chaudhry said.
The platform allows IT administra- tors to make changes as they become necessary. For instance, if an algorithm interprets a physician’s handwritten “COVID” as “could,” the IT team can update the ML model. “As we are find- ing out things that are not right, we rectify it, and once we rectify it in one area, it has a downstream effect and fixes itself all across,” Chaudhry said.
MLaaS is part of a larger effort to bring emerging technologies into the FDA’s ecosystem, he added. The next iteration of MLaaS will plug into FDA’s existing low- to no-code workflow automation tools. Then Chaudhry said he will turn
rising star
Sasha Samochina
Group Manager, Data Visualization and Infusion
NASA Jet Propulsion Laboratory
Sasha Samochina’s rare mix of technological expertise, business skills and creativity has made her
a powerful force in NASA and the broader space community. Although she recently left the Jet Propul-
sion Laboratory for a private-sector opportunity, the impact of her efforts are lasting and substantial.
For example, she played a key role in pioneering the ProtoSpace software package, which layers engineering designs on the Micro- soft HoloLens mixed-reality headset to provide immersive visualization capabilities for science missions.
Its success on missions such as the Mars 2020 Perseverance rover has laid the groundwork for operational- izing the capability as an IT enter- prise service at NASA.
Samochina’s focus on experien- tial technology that incorporates innovative forms of visualization has also made its way into content she created for communication channels and the Mars Curiosity rover. She put together the first 360-degree video released on NASA’s social media, for example. She also directs some of her bound- less energy to encouraging young people, especially girls, to pursue careers in science and technology.
44 November/December 2021

   42   43   44   45   46