What you look for online can undoubtedly hint at who you are as an individual. Now, researchers are turning to these web searches to explore how they can be used to aid in detecting cancer early.
In a study published in the Journal of Oncology Practice, a team from Microsoft showed that web searches can be used under certain circumstances to predict a future diagnosis for pancreatic cancer, using data collected from the search engine Bing.
By identifying recent search queries indicative of pancreatic cancer, such as the term, "I was told I have pancreatic cancer, what to expect," the researchers worked back to searches months earlier to examine patterns in symptoms that concerned individuals usually search for. Some of previous search terms they associated with a future pancreatic cancer diagnosis include "yellowing skin," "dark urine," "abdominal swelling" and "dark or tarry stool."
When patterns were found and analyzed, the researchers were able to identify between 5 and 15 percent of cases with low false-positive rates.
Robert Grossman from the University of Chicago, who is not involved in the study, thinks the research is exciting and a big step forward.
"Finding low cost, low risk, high coverage health surveillance systems is an important challenge," said the professor of medicine.
However, there are concerns about the results of the study. Alison Patricia Klein, an oncology professor from the Johns Hopkins University, for instance, warned that there are limitations to how beneficial using web searches for early cancer detection can be.
She said that she is concerned that those being identified will already be in the disease's later stages where not many effective therapies are available. Additionally, results from the study might not be representative enough because low-income patients are not likelier to have access to technology, and thus will be incapable of performing the web searches the researchers analyzed. Klein thinks this is a problem because risk factors may differ in different people.
There's a lot of potential in analyzing health data because there's a wider range of access points now. Aside from making web searches, individuals can take advantage of wearables that record health data specific to them. Gaining access to these data will give not just the current researchers but others as well a trove of information that can be used to improve health outcomes for everyone.
Authors for the study include: John Paparrizos, M.Sc., Ryan White, Ph.D. and Eric Horvitz, M.D., Ph.D.
Photo: Hobvias Sudoneighm | Flickr
