Slashdot Mirror
Microsoft Analyzes Web Searches, Finds Clues For Early Cancer Detection (computerworld.com)
An anonymous reader quotes a report from Computerworld: Analyzing online activities can provide clues as to a person's chances of having cancer, Microsoft researchers showed in a paper published this week. Specifically, the researchers demonstrated that by analyzing web query logs they were able to identify internet users who had pancreatic cancer even before they'd been diagnosed. The study suggest that "low-cost, high-coverage surveillance systems" can be created to passively observe search behavior and to provide early warning for pancreatic cancer, and with extension of the methodology, for other challenging cancers," the researchers concluded. "Surveillance systems could also provide for automated capture and summarization of data and landmarks over time so as to provide patients with talking points in their discussion with medical professionals." The researchers used proprietary logs of 9.2 million web queries on Microsoft's own Bing search engine but focused exclusively on English-speaking people in the U.S. from October 2013 to May 2015. First, the team identified searchers in logs of online search activity who made "special queries" that are suggestive of a recent diagnosis of pancreatic cancer. Those queries included phrases such as "Why did I get cancer in pancreas," and "I was told I have pancreatic cancer, what to expect." The team then went back "many months" before the initial queries were made to examine patterns of symptoms as they were expressed by web searches about pancreatic cancer symptoms. "We showed specifically that we can identify 5% to 15% of cases, while preserving extremely low false-positive rates," the researchers said in their paper. The false positives ranged from one in 10,000 to one in 100,000.
Actually, if you had patience to read the summary, they did something very clever. They did find the people who had been actually diagnosed with cancer. Then they went back months to their previous searches, and found that BEFORE they had any idea they had cancer, they were searching for information on their symptoms... symptoms of the cancer that would be discovered much later. How much would it be worth to you to find out you have cancer when it can be treated, rather than too late?
> If they had a false positive rate of 1% or even 10%, that would be amazingly good.
If it was a sensitive test, that'd be true.
But it's not a sensitive test. It spots a relatively small percentage of pancreatic cancer cases early, while getting a relatively small number of false positives.
Diagnostic and screening tests trade off sensitivity (detecting cases) and specificity (positive tests indicating the specific condition). This test has a very low sensitivity in order to attain its high specificity.