Siemens Healthineers and Intel Demonstrate the Potential of AI for Real-Time Cardiac MRI Diagnosis

What’s New: Intel and Siemens Healthineers* are collaborating on a break-through artificial intelligence (AI)-based cardiac MRI segmentation and analysis model that has the potential to provide real-time cardiovascular disease diagnosis. Using 2nd-generation Intel® Xeon® Scalable processors for AI inference, Intel and Siemens Healthineers demonstrated the ability to deliver MRI inferencing results to technologists, cardiologists and radiologists in real time.

“Siemens Healthineers and Intel have a shared goal to improve healthcare by applying AI where the data is generated — right at the edge using 2nd-generation Intel Xeon Scalable processors with Intel® Deep Learning (DL) Boost and the Intel® Distribution for OpenVINO™. This enables real-time applications of cardiac MRI, making data interpretation available right after it’s collected.”
–David Ryan, general manager, Health and Life Sciences Sector, Internet of Things Group, Intel

Why It’s Important: One-third of all deaths in the U.S. – 34 deaths a minute or 18 million deaths a year – are due to cardiovascular disease1. Cardiac MRI has established itself as a gold-standard for evaluating heart function, heart chamber volumes and myocardial tissue evaluation1. To extract quantitative measurements from the CMR images, the cardiologists typically use manual or semi-automatic tools, a time-consuming step that is error-prone and affected by the inter-user subjectivity in interpreting the images.

“We can now develop multiple real-time, often critical medical imaging use cases, such as cardiac MRI and others, using Intel Xeon Scalable processors, without the added cost or complexity of hardware accelerators,” said Dorin Comaniciu, senior vice president, Siemens Healthineers.

Utilizing an AI model of the heart potentially saves time for cardiologists because they do not have to manually segment different ventricles, myocardium, and blood pool cavities. AI-based segmentation happens as soon as the image slices are generated by the scanner – at the edge, where the computation system can keep pace with the data being generated. This provides low latency for AI inference and high throughput speed, enabling healthcare providers to safely increase the number of patients treated per day.

What Benefits It Offers: The health and life sciences industry is digitizing healthcare and utilizing AI to accelerate clinical workflows, improve accuracy and diagnosis, reduce hospital costs, and support medical research. AI can quickly provide visibility into anatomical systems and identify abnormalities, which helps clinicians focus patient care.

About the Technology: Most systems deployed by Siemens Healthineers are already powered by Intel® CPUs, allowing Siemens Healthineers to leverage its existing CPU-based infrastructure to run AI inference workloads. Siemens Healthineers and Intel used the Intel® Distribution of OpenVINO toolkit to optimize, quantify and execute the model. The resulting demonstration achieved a more than five times speed-up with almost no degradation in accuracy.2

Intel DL Boost is a new set of embedded processor technologies designed to accelerate deep learning use cases. It extends Intel® AVX-512 instructions with a new Vector Neural Network Instruction (VNNI), which is built into 2nd-generation Intel Xeon Scalable processors. Tasks such as convolutions, which typically required three instructions, can now be accomplished with just one instruction. Examples of these targeted workloads include image-recognition, image-segmentation, speech-recognition, language-translation, object-detection and more.

More Context: Siemens Healthineers and Intel (White Paper) | Data-Centric Innovation at Intel | Artificial Intelligence at Intel

The Small Print:

1 Journal of the American College of Cardiology, February 9, 2017.

2 Configuration: Based on Siemens Healthineers and Intel Analysis. Intel measured on Feb. 28, 2019, using 2nd-Gen Intel® Xeon® Platinum 8280 Processor with Intel® OpenVino™. Comparing FP32 vs Int8 w/ Intel® DL Boost performance on the system- 5.5x speedup. Configured with 192 GB of memory, Intel® Solid State Drive Data Center 480 GB, and CentOS Linux* 7.4.1708 kernel 4.19.5-1.el7.elrepo.x86_64; Intel® MKL-DNN, Intel® Distribution of OpenVINO™ Toolkit (R5 Release). Dense U-Net: CNN for biomedical image segmentation.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. Check with your system manufacturer or retailer or learn more at intel.com/iot.

Performance results are based on testing as of the date noted in configuration details and may not reflect all publicly available security updates. See configuration disclosure for details. No product or component can be absolutely secure.

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.

Cost reduction scenarios described are intended as examples of how a given Intel-based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does not guarantee any costs or cost reduction.

Optimization Notice: Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice. Notice Revision #20110804

Intel, the Intel logo, OpenVINO, and Xeon are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries.

This Siemens Healthineers’ feature is currently under development; and is not currently for sale.

About Intel

Intel (NASDAQ: INTC), a leader in the semiconductor industry, is shaping the data-centric future with computing and communications technology that is the foundation of the world’s innovations. The company’s engineering expertise is helping address the world’s greatest challenges as well as helping secure, power and connect billions of devices and the infrastructure of the smart, connected world – from the cloud to the network to the edge and everything in between. Find more information about Intel at newsroom.intel.com and intel.com.

Intel and the Intel logo are trademarks of Intel Corporation in the United States and other countries.

*Other names and brands may be claimed as the property of others.