NASA data scientists are helping fight cancer

In In The News by Barbara Jacoby

By: Brooks Hays


Researchers with the National Cancer Institute continue to make cancer-fighting discoveries using data analysis techniques developed by data scientists at NASA’s Jet Propulsion Laboratory.

NASA scientists are constantly trying to figure out ways to manage and mine the data being pulled in by the many instruments and sensors attached to their exhaustive catalogue of probes, satellites, observatories and arrays.

Thanks to a renewed agreement between JPL and NCI, cancer researchers will be able to keep taking advantage of the big data expertise developed by NASA scientists.

Many of NASA’s algorithms work by parsing massive amounts of data for anomalies that might be interesting to astronomers and physicists — they pull out the statistical or mathematical needles from the proverbial haystack.

Members of the NCI-supported Early Detection Research Network want to do the same thing, only their data isn’t cosmological clutter — it’s reams and reams of biomedical data.

Beginning in 2000, scientists working on the EDRN began pooling their data concerning several cancer biomarkers, as well as chemical or genetic signatures linked to specific cancers. Over the last 16 years, more and more data has been amalgamated into a single, searchable network. But the data set has grown overwhelming.

EDRN needed help standardizing their data and finding a way to analyze metrics from a variety of instruments. Their algorithms needed to be taught perspective and integration.

“We didn’t know if they were early-stage or late-stage specimens, or if any level of treatment had been tried,” Sudhir Srivastava, chief of NCI’s Cancer Biomarkers Research Group and head of EDRN, explained in a news release. “And JPL told us, ‘We do this type of thing all the time! That’s how we manage our Planetary Data System.'”

“The more we expand, the more data we integrate,” added Christos Patriotis, program director at NCI’s Cancer Biomarkers Research Group. “Instead of being silos, now our partners can integrate their findings. Each system can speak to the others.”

As scientists with NCI and NASA look ahead to the data-powered fight against cancer, new challenges await. Researchers are hoping to integrate visual recognition capabilities into EDRN’s programming.

In the future, EDRN’s machine learning algorithms could compare a patient’s CT scan to its expansive archive to zero-in on possible risks.

“As we develop more automated methods for detecting and classifying features in images, we see great opportunities for enhancing data discovery,” said Dan Crichton, head of JPL’s Center for Data Science and Technology. “We have examples where algorithms for detection of features in astronomy images have been transferred to biology and vice-versa.”