How can you map a building without going inside? Big data analytics, of course.

We are getting used to the idea of big data now that Waze knows to send us a coffee coupon as we drive near a Dunkin' Donuts. Analytics is becoming a fact of life. We know pop-up ads will match our interests.

But what happens when we want to analyze data that we haven't collected yet? This week, we hear from a Vencore computer scientist who tackled just that problem. He wanted to solve a problem with analytics, but there was no data to help him…

Tara Grabowsky MD
Chief Medical Officer

For more information, visit HVHprecision.com


I took my first dive into the world of big data during my time at the University of Pennsylvania. Charged with using our experience in engineering to solve a real world problem, my senior design team and I formulated our base question: is it possible to generate a map of any building we enter without prior knowledge of that building? The immediate answer was "no." You can't draw a map without any information about the area of interest. How could North America have been mapped without sailing the coastline? We decided to ignore the nagging naysayers who claimed "you can't build something from nothing!"

We wanted to create a Google Maps of sorts, offering newcomers the ability to find their way around the inside of the building. We needed a system that could be applied across a multitude of interior spaces with fast results. We needed to gather information about a building and then compile it into a spatially accurate map. We needed to know the difference between halls, doors, walls and rooms. We needed to know distances and dimensions. We needed data.

So we turned to the greatest source of loose data that exists: people. Whether aware of it or not, people are data goldmines, acting as constant streams of information output. And with today’s prevalence of mobile devices, people have become constant streams of readable information output.

With this in mind, we began to develop an ad hoc network of mobile devices constantly locating themselves within a building and sharing the data with a home node. Our system worked like this:  A user enters a building and begins automatically locating wireless routers via a WiFi-enabled mobile device. The user is then pinpointed to a location based on the triangulation of WiFi signal strengths measured between the mobile device and three different routers. As the user moves throughout the building, the mobile device constantly ‘locates’ itself in relation to the routers present in the building. This individual user location information is then sent to a home node, which can compile the information of many users and overlay them into a heat map of movement. Enough users over a period of time allows the home node to infer the location of halls, walls, doors, and rooms and create a 2-dimensional mapping of the building’s interior.

It was by no means an easy task, but by the end of the year we had developed our mapping system.  It was a gratifying accomplishment, but the truly lasting aspect of the project was the versatility we witnessed in the use of data to accomplish a goal. We were able to create a means of data collection, collect and study the necessary data, and apply it to a larger system to produce something creative and useful. We had, in fact, built something from nothing. We had taken a problem for which we had no data and developed a fully-functional method of mapping unknown areas, all thanks to data analytics.

Mike Fisher
Vencore Data Scientist

Lunar Ice-Trap ISRU Mining, Processing and Storage Infrastructure

Welcome back to the “Vencore Geek Tribute” series. As I delve further into the world of healthcare analytics, I marvel more about the power of cross-science dialogue. I love to sit at the table with scientists whose brains have been trained differently, and watch those disparate perspectives create new approaches to the challenges of scaling computing infrastructure, handling terabytes of data, and ensuring clinical relevance. Our last post focused on the combination of physics and biology. I am a clinician-turned-medical analytics expert. Today, we hear from one of our aerospace engineers. He has studied the intricacies of sending large objects to the moon, and now he is helping us to expand our healthcare computing platform and analytics.

Tara Grabowsky, MD
Chief Medical Officer, Vencore


In-Situ Resource Utilization (ISRU) is defined as "the collection, processing, storing and use of materials encountered in the course of human or robotic space exploration that replace materials that would otherwise be brought from Earth."

During my senior year at Penn State University I joined a team of six aerospace engineers to design a water ice mining facility in a permanently shadowed crater located in the lunar South Pole. This was theoretical, of course, but over the next nine months I felt like I traveled to the moon and back!

We broke the project into subsystems: structures, propulsion, launch vehicle, power, thermal, communications, guidance navigation and control, command and data handling, ground control, mission architecture, and scientific instruments. I designed the structures subsystem.

Each subsystem had its own set of aerospace-related theories and calculations. For example, structures was responsible for calculating required sizes of the infrastructure, what type of material to use, then calculate mass estimates of the space rated material used. Launch vehicle then used those mass estimates to choose the best launch vehicle based on our needs and mission budget. Propulsion used various rocket equations to calculate efficient propellants to use. Communications had an arsenal of equations to determine required antenna sizes and data rates.

We worked the mission from the brainstorming phase to a fully designed conceptual mission. I worked with my teammate responsible for the thermal subsystem. We integrated his mission requirements with the structures I had designed for the mission. I also collaborated frequently with the launch vehicle system to design structures and rovers to fit efficiently into the selected vehicle’s payload fairing. One challenge we had to overcome was the propellant storage unit, where the propellant we produced would be stored. A mission requirement to store 100 tons of propellant annually was quite a challenge. A standard solid metal tank is no big deal for storage on earth, however transporting it to the moon is. The storage tanks were limited by both the payload fairing of the launch vehicles as well as the mass of such a large structure. I designed an inflatable storage tank made out of flexible materials at cryogenic temperatures that could hold tons of propellant while capable of hitching an easy ride to the moon. The tanks used a network of aluminum trusses to unfold upon lunar arrival, all encased in a three-layer fabric comprised of Teflon®, Cryogel® Z and woven Kevlar®.

In the end, we had designed a resource refining facility, power and data distribution facility, mining rovers, mapping rovers, storage units and a repair rover. It was a system that was fully autonomous and able to produce 100 tons of hydrogen/oxygen propellant annually.

Though this project was theoretical, I learned practical knowledge to “launch” my career as a rocket scientist.

Tobie Sneeringer
Data Scientist

From Physics to Healthcare Analytics

We started the “Vencore Geek Tribute” series in our last post with a discussion of Genomics. Today, one of our data scientists discusses her transition from Physics to Healthcare – while studying sickle cell anemia. Stay tuned for posts from other scientists in our midst, including aerospace, electrical, and mechanical engineers, as well as computer scientists. If we’re lucky, we will even get a weather lesson from our own meteorologist.

Tara Grabowsky, MD
Chief Medical Officer, Vencore


People who know I have a Ph.D. in physics often ask me, how is it that I am now working in Healthcare Analytics?  I always find this a funny question, because to me, who traveled this path, it all just fits. 

When I first started studying physics in College, all my research projects involved a medical or biological component.  My first undergraduate research project involved studying how the optical properties of collagen solutions change under varying temperatures.  Later, I would do a study on the conductivity of thin DNA films.  In my physics classes I was learning the complex laws and theories which governed physical systems and how to use advanced mathematics to describe their behavior.  Meanwhile in my research, I was learning to apply this rigor to the study of biological systems.

In graduate school, my projects became much more complex and took years to complete instead of just a summer.  Having always been attracted to the medical field, I chose biophysics as my research concentration.  Biophysics is a growing field, where the ideas and mathematical models which have been effective in describing physical systems are applied to biological ones.  It involves the measuring and modeling of data to understand the behavior of a system and/or to predict future or unmeasurable (in the lab) behavior.

The example I like to give of biophysics is my own dissertation research.  My thesis work was on the disease Sickle Cell Anemia.  In particular—the underlying cause of the disease: the aggregation of the hemoglobin in the red blood cells which stiffens the cell and occludes blood flow.  The aggregation is triggered when the red blood cells deoxygenate.  My experiment involved triggering this event in samples and measuring it under controlled and varying conditions.  With the results, I produced a model which defines with what probability red blood cells are sickling in a patient during transit under varying physiological conditions. The motivation for the work was to provide a model of Sickle Cell Aggregation to aid other researchers in their development of a therapy: Link.

What I have learned from my work is how to model a complex and chaotic biological system using the principles of physics and the application of mathematics.  My physics background has given me experience in studying and analyzing data, as well as using the tools for doing so.  As an outsider, I often find that I look at the data and models in a different way from those who are trained in data analysis fields.  This has proven invaluable time and again, especially on the most difficult and seemingly impossible problems.

I dedicated many years of my life to studying a rare disease during my thesis work.  It had long been my goal and motivation to bring something of value to the study of Sickle Cell Anemia Disease.  Now, at Vencore Health Analytics, I have the opportunity to bring something of value to the study of many rare diseases.  It has been a long journey, but I am exactly in the right place and it was my physics degree which helped me get here.

 Donna Yosmanovich, PhD
Data Scientist

 

Genomics Offers Powerful Data Set to Improve Drug Development

We have been blogging for the past few months about rare disease. We have heard stories of resilience, of acceptance, and of heroics. These patients get us out of bed every morning; they are part of our MISSION. But what enables us to help patients with rare disease? We are, at the end of the day, a company of science. At Vencore we have over 500 data scientists and 180 Ph.D.s in every field from astrophysics (so yes, it actually IS rocket science) to mathematics to electrical engineering. We have imaging scientists, computer programmers, and even our very own meteorologist. So this summer we plan to use those skills to launch a series on data science, and all things technical. We start today with genomics. The data being generated in this field are staggering. How will we manage it all? How will we use it in a way that is clinically intelligent? Read on for a brief introduction to the topic from one of our data scientists.

Tara Grabowsky, MD
Chief Medical Officer, Vencore


Drug development is a risky business. Recent articles show that the risk-adjusted cost of bringing a new drug to the market can add up to more than $2.5 billion. This is an astonishing amount considering that about only 10% of these drugs survive through the clinical development process. This imbalance between cost and returns has resulted in biopharmaceutical companies modifying their approach to change the cost-risk equation and enhance returns. One such modification is using genomic data to make informed decisions about drug development, which can ultimately lead to higher efficacy.

Genomics is the study of genome sequencing and analysis. The information it provides can be used to understand and explore the many expressions of living organisms including identifying disease risk, ancestry, traits, response to medicines, drug targets and validation. It is known that the genome is 99.9% identical among the human species (and up to a staggering 98% identical between humans and other species), yet somehow there are millions of differences in the genomic expression of individuals. Understanding why and how the genome is expressed differently on an individual basis allows us to understand why some people respond to one medicine better than another, or are at risk for certain diseases. This allows us to be more precise in our approach to healthcare.

Through the understanding of genomic information, pharmaceutical companies now have the potential to produce targeted therapies that affect only the disease agents and leave healthy cells untouched. For example, breakthrough discoveries such as the recent advent of CRISPR/Cas-9 system, through functional genomics screening, can help identify novel targets that were once overlooked. This discovery also aids in simplifying disease models by using permanent gene knockouts to validate targeted therapeutics.

Pharmaceutical companies have already begun incorporating genomics information into their programs. Last year some of the world’s largest pharmaceutical giants such as GlaxoSmithKline, Astra Zeneca, Roche, AbbVie and Biogen teamed up to mine the genomes of 100,000 patients with cancer and rare diseases. This year AstraZeneca has started an initiative, partnering with strong genomics leaders, to mine up to 2 million people’s genomes for new drug targets. This is an unprecedented amount of information that will be available as a new resource. They have already established an in-house genomics center and will start another research team in the United Kingdom’s Wellcome Trust Sanger Institute.

In this era of personalized medicine where genomics has quickly taken place as a key player, it becomes increasingly crucial to be knowledgeable of the advantages genomics information can provide to the healthcare industry. More importantly, it is necessary to know how to interpret this data so we can fully understand and integrate with other medical information to build more holistic pictures for individual patient care.

Manjula Kasoji, MS
Data Scientist

Rain Can’t Stop the Fight Against Rare Disease at the Million Dollar Bike Ride

Chris Miller, program manager, Vencore at the Million Dollar Bike Ride. 

Chris Miller, program manager, Vencore at the Million Dollar Bike Ride. 

We are in the business of connecting rare disease patients to therapy. We are also always seeking ways to be involved in the rare disease community that go beyond our business goals.  This past weekend offered one such opportunity at one of the biggest local rare disease events of the year: the 3rd Annual Million Dollar Bike Ride sponsored by the Penn Medicine Orphan Disease Center (ODC).

Despite being a relatively young event, the Million Dollar Bike Ride has annually brought over 600 participants from more than 20 U.S. states and has raised over $2 million for rare disease research in two years. Perhaps even more incredibly, it is reported that 100% of the funds raised have gone directly to rare disease pilot grant programs, without any overhead withheld.

Even though I can’t claim to be an avid cyclist, I headed down to University City on a Saturday morning to experience the event and lend some moral support to some of our friends who were riding. While the riders challenged themselves to complete rides of 12, 33, or even 73 miles, I figured just being awake and at the race at 6 a.m. on a Saturday would be challenge enough for a nocturnal person like myself.

With a 24-ounce coffee in hand, I made it to the Penn ice rink as the cyclists were gathering.  The event said rain or shine, and the organizers definitely weren’t kidding. The weather was miserable enough for standing outside drinking coffee, let alone riding a bike from West Philly up the hill through Mt. Airy back along the Schuylkill to the finish (mercifully the Manayunk Wall was absent from the route, because that would have been downright sadistic in any weather). However, for people who spend every day of their lives fighting rare disease, a little rain on a bike ride is nothing to overcome. So many riders, including patients, patients’ families, and advocates, still turned out to ride for more rare diseases than I could count.

The more I interact with the rare disease community at events like the Million Dollar Bike Ride the more amazed I am at the energy the community pours into fighting rare disease.  Congratulations to all the riders and volunteers on another successful event. I look forward to coming back next year…as a participant!

Chris Miller
Program Manager, Vencore