Martin J. Bergman
MAUI, Hawaii — Protecting the large volumes of medical data, including patient data, has become an increasingly important challenge in the current medical landscape, according to a presentation at the 2020 Rheumatology Winter Clinical Symposium.
“Conventional ways of handling [patient data] with statistics do not work anymore,” Martin J. Bergman, MD, FACR, FACP, clinical associate professor of medicine at Drexel University School of Medicine, and chief of the division of rheumatology at Taylor Hospital, told attendees. He referred to the “three V’s” of big data: variety and volume of data coming at an increasing velocity.
Bergman warned that the key to big data is “capital.” Meaning, companies like Google and Facebook are making money by sharing, or selling, people’s information to interested parties. “This is what is driving most of our large tech companies, so this is something we have to understand, not only in medicine, but in our day-to-day lives,” he said.
There is a broad range of sources of medical big data, including insurance claims, registries like CORONNA or the National Data Bank of the Rheumatic Diseases, electronic medical records, social media and wearable technologies. Moreover, the ability of the large tech companies to read and interpret data is becoming more sophisticated. “It is no longer just numbers that they can see, but they can read words,” Bergman said. “For example, the RISE (RNA Interactome from Sequencing Experiments) database can read the difference between synovitis and no synovitis.”
Protecting the large volumes of medical data has become an increasingly important challenge in the current medical landscape, according to Bergman.
Source: Healio Rheumatology
Bergman then raised a pivotal question for attendees to consider. “Do you know who owns your patients’ data?” he said.
With that, the talk turned to ways of protecting these data. “One of the approaches is block chain computing,” he said.
The process of block chain computing begins with a producer creating the data using password protection. “Let’s say this is the data for a patient, who has their own password, and a known identifier,” Bergman said. “You as the provider also have a password, so the two of you can add or exchange information about your patient visit. This is called a node.”
The information is then encrypted and sent to the network in a so-called “block.” These data undergo analysis from multiple sources to determine if it meets validation requirements. If the data are validated, it is added to existing blocks of similarly protected data to form a block chain. The chain is then locked for modification and becomes the basis for further block chains. “This information is inviolable, and cannot be modified by anyone,” Bergman said.
There are two key drawbacks to this system, according to Bergman. One is that it requires large amounts of data storage from any given site or hospital system. The other is that once the data are locked, it can be difficult to correct errors. “How many times have you had a patient with a positive ANA and neck pain, and the final diagnosis is lupus, and now it is encrypted in their data forever?” he said. “I have not seen a way to get that out of their block chain.”
What is clear is that there is an increasing amount of confidential data available, and clinicians should be aware of ways to protect both themselves and their patients, according to Bergman. – by Rob Volansky
Bergman MJ. Hot Topic: Blockchain, AI, Big Data, etc: Rheumatology perspective. Presented at RWCS Annual Meeting; Feb. 12-15, 2020; Maui, Hawaii.
Disclosure: Bergman reports consultant, speaker, and/or honoraria from AbbVie, Amgen, Boehringer Ingelheim, Celgene, Genentech, Horizon, Merck, Novartis, Pfizer, Regeneron, Sanofi and Sandoz, as well as stock options in Johnson & Johnson.