Cancer Data Science Pulse

Living on the Edge—How Video Games Helped Shape the Future of Cancer Care

Given the title of this blog, you might be wondering, “What could video games possibly have to do with the future of cancer research and cancer care.” In reality, there’s much more of a connection than you might think.

You’re likely familiar with video games in one form or another. Maybe your memories take you back to the arcade games of Pong or Asteroids. You might be playing the latest version of Madden NFL on your game console at home, or maybe even Candy Crush on your smartphone. Whatever the game, they all have something in common—computer models.

What’s a computer model? To put it simply, a computer model is the “brain” behind the game. The computer model takes your inputs and converts them into action, ultimately providing the gaming experience that you see on your screen. Video games have advanced over the years. Today’s games offer nearly unbelievable levels of realism, ushering in truly amazing visual detail and virtual reality.

The Cancer Connection

Thanks, in part, to video games (and a lot of other technologies), you now can connect to the Internet and play against other gamers across the globe. The “edge device,” whether it’s a computer, a video game console, or even a mobile phone, makes this possible—it’s the “edge” (i.e., the place) where players meet and interact with the game and with one another. “Edge computing” is what makes it possible for the device in your hands to process your inputs and provide the outputs for a total gaming experience, including those complex realistic graphics from a graphics processing unit (GPU).

But orchestrating a winning touchdown on a virtual football field is one thing. Where’s the cancer connection?

The same basic computer technology that’s adding to the growing realism of video games also powers important advances for cancer research. The General Purpose GPU (GPGPU) has become central to the rapid growth of artificial intelligence (AI) and use of machine learning to create data-driven models and even large language models (LLMs). Data scientists are using this GPGPU technology every day to:

  • analyze large volumes of data,
  • identify features and trends,
  • form relationships,
  • build connections, and
  • optimize accuracy.

Using this information, we’re able to develop increasingly capable computer models.

Taking Data to the Edge

Today, we have a tremendous volume and variety of data available to study cancer—genomic, proteomic, imaging, biometric, clinical trials, and more. NCI’s CBIIT provides one such source of data with NCI’s Cancer Research Data Commons. Using those data, we’re creating models to predict, track, and monitor cancer in ways we never could before. For example, it’s possible to now use routine pathology images and genomic data to identify important features in a tumor to help select treatment with the greatest expectation of success—what’s known as precision medicine, and in the case of cancer, known as precision oncology.

We’re seeing edge computing’s role grow as new algorithms and computer models develop and move into clinical workflow, offering insight into cancer and supporting decisions for individual patient care. Using this technology, we’re able to collect key longitudinal information about individuals both during treatment and during follow-up. We’re also exploring the use of edge computing to help solve problems in latency, bandwidth, and costs associated with working in the cloud.

NCI’s Efforts in Edge Computing

At NCI, our researchers are pioneering the use of edge computing. Here are just a few examples:

Next Steps—Pushing the Edge

Perhaps edge computing’s greatest advantages will be in a clinical setting. More and more clinicians and people with cancer are using these technologies. One device with tremendous potential is the now-common smartphone, which is increasingly serving as a personal health data hub.

For instance, people with cancer can use their smartphones as input devices, sharing insight with their doctors on how they feel at any time, night or day, and receiving feedback and information. These devices also can connect to new “biometric” sensors, which monitor a number of physiological events, such as physical activity, heart rate, and body temperature. Such data could be useful in monitoring individual responses to treatment and making adjustments to optimize outcomes.

These are only a few examples of how we can apply edge computing to clinical practice. Using edge computing, we can directly connect doctors and their patients, creating a two-way learning system for cancer. Most importantly, we’ll be able to share data in a way that not only helps one patient but benefits every person living with cancer.

In future blogs, we’ll look more at how these exciting emerging data science topics are shaping the future for cancer research. For example, in the area of digital twins, we’re using some of the largest computers in the world to advance cancer research and improve patient care, as detailed in a recent report from the National Academies of Sciences, Engineering and Medicine, co-sponsored by the Department of Defense, the Department of Energy, NIH, and the National Science Foundation.

We’d welcome your feedback on blog topics you’re most interested in seeing. Please take a moment to comment in the box below.
Eric Stahlberg, Ph.D.
Director, Cancer Data Science Initiatives, Cancer Research Technology Program, NCI’s Frederick National Laboratory for Cancer Research
Older Post
The 12 Blogs of CBIIT—Your Favorite Data Science Reads from 2023
Newer Post
Why We Love Diverse Data

Leave a Reply

Vote below about this page’s helpfulness.

Your email address will not be published.