SANTA CLARA, Calif.–( BUSINESS WIRE )–December 5, 2022-
Intel Labs and the University of Pennsylvania’s Perelman School of Medicine (Penn Medicine) have used federated learning—a distributed machine learning (ML) artificial intelligence (AI) approach—to help international healthcare and research institutions detect malignant brain tumors. The largest medical federated learning study to date with an unprecedented global dataset examining 71 institutions across six continents, the project demonstrated its ability to improve brain tumor detection by 33%.
This press release features multimedia. Check out the full release here:
Using Intel Federated Learning Technology combined with Intel Software Guard Extensions (SGX), the researchers were able to address numerous data privacy concerns by keeping data holders’ raw data inside the compute infrastructure and only allowing model updates computed from that data to be sent to the center. The server or aggregator, not the data itself. (Credit: Intel Corporation)
-Jason Martin, Principal Engineer, Intel Labs
Data accessibility has long been an issue in healthcare due to state and national data privacy laws, including the Health Insurance Portability and Accountability Act (HIPAA). Because of this, it is nearly impossible to share medical research and data without compromising patient health information. Intel’s federated learning hardware and software adhere to data privacy concerns and preserve data integrity, privacy and security through confidential computing.
The Penn Medicine-Intel result was accomplished by processing high volumes of data in a decentralized system using Intel Federated Learning Technology paired with Intel® Software Guard Extensions (SGX), which removes data-sharing barriers that have historically prevented collaboration in similar cancers and diseases. did The research system addresses numerous data privacy concerns by placing the raw data inside the data holders’ computational infrastructure and allowing only model updates computed from that data to be sent to a central server or aggregator, not the data itself.
“All the computing power in the world can’t do much without enough data to analyze,” said Rob Enderle, principal analyst, Enderle Group. “This inability to analyze data already captured has significantly delayed the huge medical breakthroughs that AI promises. This federated learning study shows an effective way to advance AI and realize its potential as the most powerful tool to fight our most difficult diseases.”
The senior author, assistant professor of pathology and laboratory medicine and radiology at the Perelman School of Medicine, said, “In this study, federated learning demonstrates its potential as a paradigm shift in securing multi-institutional collaboration by enabling access to the largest and most diverse dataset of glioblastoma patients ever. Considered in the literature, when all data is kept within each institution at all times. The more data we can feed into machine learning models, the more accurate they become, which in turn can improve our ability to understand and treat rare diseases like glioblastoma.”
To advance disease treatments, researchers must access vast amounts of medical data—in most cases, datasets that exceed the threshold that can produce a benefit. The research demonstrates the effectiveness of federated learning at scale and the potential benefits the healthcare industry can realize when multisite data silos are unlocked. Benefits include early detection of disease, which can improve quality of life or prolong patient life.
The results of the Penn Medicine-Intel Labs study were published in the peer-reviewed journal.
In 2020, Intel and Penn Medicine announced the use of federated learning to improve tumor detection and treatment outcomes for a rare form of cancer called glioblastoma (GBM), the most common and deadly adult brain tumor with a median survival of just 14 months. standard treatment. Although treatment options have expanded over the past 20 years, overall survival rates have not improved. The research was funded by the National Cancer Institute of the National Institutes of Health.
Penn Medicine and 71 international healthcare/research institutions have used Intel’s federated learning hardware and software to improve borderline detection of rare cancers. A new state-of-the-art AI software platform called Federated Tumor Segmentation (FeTS) was used by radiologists to define the boundaries of a tumor and improve detection of the “operable region” or “tumor core” of the tumor. Radiologists annotated their data and used Open Federated Learning (), an open-source framework for training machine learning algorithms, to run federated training. The platform was trained on 3.7 million images from 6,314 GBM patients across six continents, the largest brain tumor dataset to date.
Through this project, Intel Labs and Penn Medicine developed a proof of concept for using federated learning to derive knowledge from data. The solution could significantly impact healthcare and other fields of study, particularly among other types of cancer research. Specifically, Intel created the OpenFL open source project to enable customers to take real-world cross-silo federated learning and deploy it confidently on Intel SGX. In addition, the innovative FeTS initiative was established as a collaborative network to provide a platform for ongoing development and encourage collaboration with the FeTS platform and Intel’s OpenFL open source toolkit, both available on GitHub.
| | (Video) | (Case Study) | (quotation sheet)
Intel (Nasdaq: INTC) is an industry leader, creating world-changing technologies that enable global progress and enrich lives. Inspired by Moore’s Law, we continually work to advance the design and manufacturing of semiconductors to help our customers meet their greatest challenges. By embedding intelligence in the cloud, network, edge and every type of computing device, we unlock the potential of data to transform business and society for the better. To learn more about Intel’s innovations, visit and .
© Intel Corporation. Intel, the Intel logo and other Intel marks are trademarks of Intel Corporation or its subsidiaries Other names and brands may be claimed as the property of others.
See the source version at businesswire.com:
Contact: Laura Stadler
Keywords: California, United States, North America
Industry Keywords: Software General Health Mobile/Wireless Internet Hardware Electronic Design Automation Data Management Apps/Application Technology Hospital Semiconductor Sec
Source: Intel Corporation
Copyright Business Wire 2022.
PUB: 12/05/2022 11:00 AM/DISC: 12/05/2022 11:02 AM