FAMHE: New Federated Analytics System Could Be the Game-Changer for Personalized Medicines
The healthcare industry is seeing new changes with the help of technological innovations.
Some in the healthcare industry aim to adopt the concept of P4 medicine (predictive, preventive, personalized, and participatory). To improve the industry and its services, the clinical data of every individual must first be analyzed.
Based on the clinical data, the industry can make the required changes. Clinical data refers to the data of every individual who visits clinics or medical facilities.
This further strengthens the production departments of the manufacturers of different medicines.
Gathering individuals’ data is complex, so the industry must use a centralized data analytics system.
Sharing patients’ personal information invites additional risks. The patients’ sensitive information worries them the most.
The Solution
Researchers from EPFL’s laboratory have devised a solution to overcome this issue.
The laboratory for data security works with a team from Lausanne University Hospital, the Broad Institute of MIT and Harvard, and MIT CSAIL. The team has developed a new analytics system named “FAMHE”, a federated analytics system.
This system enables healthcare providers to collaboratively design machine learning models for statistical analysis without sharing their patients’ datasets.
What makes FAMHE a Game-changer?
FAMHE is designed to eliminate the need to share patient data, addressing major privacy concerns. The system is designed to safeguard the privacy, the accuracy of the research results, and the automatic computation of the data collected by the system.
Privacy and other accuracy changes in the healthcare industry are a matter of discussion among researchers. FAMHE and similar models aim to address these issues, and FAMHE is emerging as a highly effective system for data analytics in the healthcare industry.
The team of researchers has also released a post regarding their new Federated Analytics System and its outcomes.
The research data shows that this system can achieve scientific results without collecting datasets directly from healthcare providers. The study reports claim to have achieved the desired results through the received data.
Jean-Pierre Hubaux, EPFL Professor, Study’s Lead Senior Author has released a statement regarding their new system i.e. FAMHE. The studies show that their model works at scale. The official statement is embedded below:
Until now, no one has been able to reproduce studies showing that federated analytics works at scale. Our results are accurate and obtained with a reasonable computation time.
FAMHE uses multiparty homomorphic encryption, which allows computations on data in its encrypted form across different sources without centralizing the data or allowing any party to see the other parties’ data.
“This technology will not only revolutionize multi-site clinical research studies but also enable and empower collaborations around sensitive data in many fields such as insurance, financial services, and cyber defense, among others.”
Most universities and hospitals avoid sharing personal patient information due to privacy concerns. FAMHE enables easier collaboration and model creation using the latest artificial intelligence technology without compromising data privacy.
According to the University of Lausanne, patients want to share their personal information with such institutes; however, sharing such patients’ personal information is not a good move without data confidentiality.
The newly developed FAMHE system addresses this issue by enabling collaborative research without direct data sharing.
“This is a game-changer in personalized medicine because, as long as this kind of solution does not exist, the alternative is to set up bilateral data transfer and use agreements, but these are ad hoc, and it takes months of discussion to ensure the data is going to be adequately protected when this happens.
FAHME provides a solution that makes it possible once and for all to agree on the toolbox to be used and then deploy it”, says Prof. Bonnie Berger of MIT, CSAIL, and Broad.
The FAMHE system is still in its testing phase, and it requires further study and development through additional experiments to realize its potential as a complete analytics system.
To develop FAMHE at scale, the doctors and researchers are in talks with partners in various states across the United States along with some countries where the healthcare industry is running smoothly.
Follow us