Doctors may have found a new symptom of lung cancer — web searches.
In ongoing efforts to establish early detection techniques, Microsoft researchers believe that your search history just might serve as an interesting tool. As per a new study published in the journal JAMA Oncology, “Pattern analysis and recognition based on search log data holds opportunity for identifying risk factors and for framing new directions with the early detection of lung carcinoma.” That is to say, what you’re Googling could help doctors detect the deadly disease up to a year sooner than current diagnostic methods allow in more than one-third of cases.
As Bloomberg reports, researchers Ryen White and Eric Horvitz looked into anonymous searches in Microsoft’s search engine Bing. In particular, they examined searchers whose queries seemed to suggest a recent lung cancer diagnosis, either looking for questions about treatments or the more straightforward, “I was just diagnosed with lung cancer.”
Afterwards, they looked at these searchers’ older queries to look for patterns that might’ve suggested symptoms of cancer before the official diagnosis, such as queries into bronchitis, chest pain, or blood in sputum. Other risk factors, like gender, age, race, and environment, were also taken into consideration. Finally, the researchers looked into whether the searcher was a smoker by keeping tabs on queries related to quitting smoking or Nicorette gum.
Apparently, this method proved quite effective. “With one false positive in 1,000, 39 percent of cases can be caught a year earlier, according to the study,” Bloomberg reported. “Dropping to one false positive per 100,000 still could allow researchers to catch 3 percent of cases a year earlier.”
And when it comes to cancer, finding the problem early is key to treatment. “I have very stark memories of my first lung cancer patient as a medical student. I remember saying is this operable? He had just come in, but the lung cancer was already in his brain,” Horvitz said. “The prospect we could save patients’ lives’ with one or more methods of pre-screening is intriguing.”
Of course, the next step for the Microsoft team is actually implementing their new discovery. Horvitz notes that one potential route forward would be to create an app that monitors patients’ search history and alert them if they seem to be at risk. Horvitz and Gray are also hoping to find volunteers who are willing to grant researchers access to both their medical records and search logs so that researchers can find a stronger link between the two.