Could Online Search Help To Detect Cancer Early? These Microsoft Researchers Think So

Billions of searches are carried out daily on the internet on different subject and in this day and age, the term Google is not just used to identify the Search Engine giant but has fast become a verb. So to prove a point, we say something like “Google it”. With these searches comes an online pattern as well which is all Big Data and that’s how Google and its likes know what to show to you based on your regular searches. This kind of data could come in handy when applied under the right circumstances.

It’s been nearly two years since Google announced that it would launch a medical service so the millions of users who make health related searches are  directed to the right people online. This it hoped to achieve by connecting searchers to doctors who could be of help.

This article though is about researchers at Microsoft are using Bing (Microsoft’s search engine) search data to carry out an early diagnosis of diseases like cancer in a post earlier today. As argued by health professionals, early detection of many of diseases could be the best cure especially or diseases like cancer and that’s what this research is all about especially in a period when it’s much easier to access information over the internet with respect to health than actually going to the hospital. They said “People are being diagnosed too late,” he said. “We believe that these results frame a new approach to pre-screening or screening, but there’s work to do to go from the feasibility study to real-world fielding.”

The research was carried out by Microsoft researchers Eric Horvitz and Ryen White, along with former Microsoft intern and Columbia University doctoral candidate John Paparrizos.

In their work, they argued that  “… signals about patterns of queries in search logs can predict the future appearance of queries that are highly suggestive of a diagnosis of pancreatic adenocarcinoma, – the medical term for pancreatic cancer…we show specifically that we can identify 5 to 15 percent of cases while preserving extremely low false positive rates” of as low as 1 in 100,000.”

IBM had last year through it’s Big Data analytic tool Watson partnered with 14 leading cancer institutes to advance genomics research resonated strongly. IBM’s cognitive computing system has been participating in – Watson has dipped its toes into everything from education to cooking — this new genomics initiative marks a way to leverage big data to simplify an often complex process, such as making personalized cancer diagnoses.

According to IBM, the genome for one patient equals about 100 gigabytes of data. It’s a vast amount of information for a human doctor to sift through. With Watson quickly analyzing this information, IBM executives view this as being one of the cornerstones of the broader Watson Health initiative.

As of three years ago according to the World Health Organisation (WHO), Nigeria has the highest cancer rate by population in Africa. Cancer killed 7.6 million people worldwide in 2008 and there are indications that the figure could be 133 million by 2030 and these figures means 13 percent of global deaths are caused by cancer. In Nigeria according to the world health body, about 10,000 Nigerian die from the disease annually with some 250,000 new cases annually. Cancer research funding is increasing as the death  from it continues to rise globally.