Can Your Data Predict Disease?
Photo Source: Flickr
Every time you use the internet, your smartphone, or make a request of Alexa, you generate data. For example, when using the internet through Google’s browser Chrome, Google pays close attention to everything you do. From what you shop for, search, and read, Google and other companies will take that information and tailor advertisements to individual users. Such information travels quickly with customized ads almost immediately appearing after browsing or purchasing something online. Your browsing information clearly appeals to advertisers, and Google makes money selling those ads. Beyond commerce, your data can also be used in other ways, and researchers have found that multiple streams of data can, with the help of artificial intelligence, impact human health by predicting disease.
Early detection of disease or prediction of infectious disease outbreaks can lead to preventative treatments and better preparedness for healthcare providers. Highly contagious influenza travels very quickly and debilitates many people and can even prove deadly in the old and sick. Google found that monitoring people’s searches of flu-like symptoms helped detect in near real time the flu outbreak, but it could not predict flu outbreaks. Nancy Fliesler wrote on the Boston Children’s Hospital Innovation blog Vector a post titled, “Using multiple data streams and artificial intelligence to ‘nowcast’ local flu outbreaks.” In the post, she describes an approach by researchers at the hospital in the Computational Health Informatics Program that uses artificial intelligence combined with the Google data, electronic health records from hospitals all across the country, and historical data. Their work has been spectacular. They demonstrated last year the ability of predict flu outbreak a week in advance in three quarters of states in the US. Such lead time should help health authorities to ramp up prevention campaigns.
At a more personal level, data and artificial intelligence show signs of better predicting diseases such as heart failure and Alzheimer’s disease. An article in Science Daily titled, “Using AI to detect heart disease,” describes research at the University of Sothern California that uses an iPhone and artificial intelligence to measure arterial stiffening or hardening of the arteries. Arterial stiffening begins before diseases such as heart attack, stroke, and kidney failure. Using just an iPhone, individuals can monitor their pulse waves. The camera on the iPhone captures the pulse in the familiar wave form, and their AI then analyzes the form and can determine if the individual has arterial stiffening. The author points out that this type of analysis before the iPhone app needed to be done with an $18,000 device in a doctor’s office. Now such a measurement can be done anywhere by anyone for little cost. Alzheimer’s Disease can have devastating effects on individuals and their families lives. The disease robs people of their memory and leads to dementia. Researchers believe that with earlier detection of this devastating disease, which can begin years before memory loss appears, medicine and life style changes could be applied to stave off the disease. In an article titled “Researchers use AI to detect early signs of Alzheimer's,” in Medicalxpress.com, the author details the use of a routine blood test that measures blood components such as vitamin B12 and hormone levels and machine learning combined disentangle a very complex set of data. They found a number of elements such as low B12 and several other markers of cardiovascular disease and immune markers combined to provide earlier detection of Alzheimer’s.
The constant use of our digital devices from computers on the internet to smartphones for communication, shopping, and navigation generates massive amounts of data. Your browsing history and purchases get deployed to tailor advertising to individuals at a massive and continuous scale. Such volumes of data combined with artificial intelligence also proves very valuable in the field of predictive and preventative healthcare. Researchers have developed AI programs that now can predict flu outbreak up to a week in advance, the risk of heart failure from images taken by the camera on an iPhone, and have gained a foothold in the important field of early Alzheimer’s detection. Preventative medicine will continue to develop and improve with more data and improved artificial intelligence.
Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.
You can buy his book on Amazon in paperback and in kindle format here.