How to protect privacy while sharing treatment information to speed up research

In one year, Luo Ji Siwei's New Year's Eve speech mentioned a case in which artificial intelligence learned within a day the medical files that humans took 140 years to read, and developed a treatment plan for a patient with intractable cancer and saved the patient— -The audience was deeply shocked

The intersection of artificial intelligence and healthcare is exciting

By analyzing large data sets and finding patterns in them, almost any new algorithm has the potential to help patients-artificial intelligence researchers only need to access the correct data to train and test these algorithms

The Security Artificial Intelligence Laboratory (SAIL) is solving these problems with a technology that allows artificial intelligence algorithms to run on encrypted data sets that will never leave the data owner's system

SAIL co-founder and Massachusetts Institute of Technology (MIT) professor Manolis Kellis hopes to save researchers from negotiating with hospitals to obtain data to run machine learning algorithms

Unleash the full potential of AI

As an undergraduate student in computer science and molecular biology at MIT, Kim worked with researchers from the Computer Science and Artificial Intelligence Laboratory (CSAIL) to analyze data from clinical trials, genetic association studies, and hospital intensive care units

In order to participate in SAIL's plan, hospitals and other health care organizations set up a node behind the firewall to allow researchers to obtain some of their data

This approach allows a wider range of researchers to apply their models to large data sets

Kellis said: "We invite machine learning researchers to come here, train based on last year's data, and predict this year's data

By anonymizing large data sets, SAIL's technology also allows researchers to study rare diseases-in these diseases, small collections of relevant patient data are often scattered across many institutions, which makes it difficult for artificial intelligence models to apply to the data

"We hope that all these data sets will eventually become available," Kellis said

Make the future of medicine possible

In order to process large amounts of data on specific diseases, SAIL is increasingly seeking to cooperate with patient associations and alliances of healthcare organizations, including an international medical consulting company and the Kidney Cancer Association