Using personal data
Singapore Telecom's research is not the only example of researchers using data to track diseases.
Researchers at Boston Children & amp; s hospital had reasons to be concerned a week before the outbreak of Ebola virus was confirmed in Guinea in 2014. On March 14, their web crawling project "health map" found eight deaths caused by "mysterious hemorrhagic fever" in Macenta, on the southern tip of Guinea. Eight days later, the Republic confirmed a total of 59 deaths associated with Ebola virus.
Founded in 2006, healtmap, a public data health initiative, demonstrates the power of data in predicting and tracking diseases, all relying solely on Web links. It has successfully tracked steam disease, dengue fever, malaria and various other viruses, rashes and diseases transmitted by mosquitoes or mice. While using data from the Internet has done a good job, researchers believe that smartphone data can help them change the course of a rapidly developing disease that is occurring.
"We do use mobile phones all the time. It can track not only the speed of our movement, but also the heart rate based on the device you have, "says Yulin hwen, a researcher at healthmap. "If you have a specific app, it can record what you're doing. People also record what they eat. They use mobile phones to shop and do banking. "
"You can get all of a person's health and social records, all of the data coming from their mobile phones," said hswin. With this approach, scientists can build personal predictive health profiles, assess a person's likelihood of getting sick, and stop the spread of the disease in the first place. If you know who is susceptible, you can guide them to take preventive measures, such as vaccines.
Privacy issues
But even handing over the data in the name of public health is quite complicated. Despite the increasing adoption of privacy regulations, such as the general data protection regulations in Europe, there is still no framework for how to share the data and who will manage it. At present, there is no good way for people to collect and share their own data, and people are becoming more and more cautious about sharing their own data.
Vasant DHAR, a professor at New York University's Stern School of business and a researcher in artificial intelligence, says some countries are working on systems that can more easily release data for such projects. For example, a data protection bill under discussion in India will create an application in which people can collect and view their own data. It will also appoint a "data trustee" for each account, who will act as gatekeeper for personal data. Data trustees, just as financial trustees deal with personal assets, release data only when it is in the best interests of the individual - perhaps for public health projects. Such a system would make it incumbent on companies and organizations to share cell phone data, so it must be transparent and secure, and even a public disease tracking program must prove its value.
"You can always build an apocalypse scenario, and obviously you shouldn't ask questions, just save the planet and make the damned data available so we can continue to live, but I think what's the point here?" DHAR said
Although mobile location data is very effective for recording real-time travel, it also has limitations. As the MIT research report points out, the quality of mobile phone data depends on its coverage. In essence, MIT can't track anyone who isn't Singapore Telecom. However, the researchers say that putting mobile data together with accurate public health data, such as census records, seems to correct the problem.
There is also the issue of anonymity. In MIT research, Singapore Telecom provided anonymous data for researchers. But the ability to make data truly anonymous remains to be seen. Cell phone data can identify where you spend your time - your office, your home, where you shop - and researchers have shown that it's relatively easy to associate your name with these details. So far, Facebook has been slow to share large amounts of data with an organization dedicated to analyzing how the platform affects elections, citing privacy and security as key reasons for its slow pace.
But this concern about data privacy comes after a few blunders. One of Facebook's mistakes was a 2014 study that tested the platform's ability to manipulate the emotions of nearly 700000 users without their consent. In the years since then, the platform has often failed to protect user data.
Lack of transparency
Today, the same companies, which have built up a lot of businesses, amass personal data for advertising purposes, and are often indifferent to data leaks and other abuses, are working on agreements with healthcare providers. This is largely controversial because they often do it in the dark, without professional confidentiality knowledge.
In 2016, Google's Artificial Intelligence Lab deepmind established a partnership with the UK's national health service (NHS), but failed to alert patients that it was getting their records for the project. In a recent debacle, Google worked with arsenson, a network of health centers, to develop tools for its doctors. Similarly, neither side has been transparent about their relationship or Google's ability to access patient data.
Governments have different histories of protecting consumer data, which could make Massaro's vision of sharing data with political decision makers unacceptable. In 2015, a hacker attack against the Federal Office of personnel management exposed the records of 21.5 million people, including their home address, telephone number, social security number and other highly sensitive information. And government agencies are also interested in spreading personal information. This year, according to foreign media reports, the California Department of motor vehicles makes $50 million a year by selling customer data to third parties such as Experian and lexisnexus.
Given the various ways in which personal data is improperly processed, researchers are now concerned that people are losing trust in academic and health institutions and that it is unlikely that such information will be provided voluntarily in the future. But unlike private companies, researchers have strict rules and ethics in protecting user data, Massaro said. "If private companies have your data and can make money from it, why can't we use it forever?"
DHAR thinks this is the wrong model and points out that Google doesn't necessarily have the personal data it has. "That doesn't mean it's right or (researchers) should use it," DHAR said.