Intel and Siemens Healthineers are working together to develop a groundbreaking ai-based cardiac MRI segmentation and analysis model that promises real-time cardiovascular disease diagnosis.Intel and Siemens healthcare used a second-generation Intel xeon scalable processor for ai reasoning, providing real-time magnetic resonance imaging (MRI) reasoning results to technologists, cardiologists and radiologists.
David Ryan, general manager of life sciences and health at Intel's Internet of things division, said: "Siemens healthcare and Intel share a common goal - to further improve healthcare by leveraging artificial intelligence technology."By deploying the Intel deep learning acceleration and Intel Distribution of OpenVINO toolkit on the edge, the second generation of Intel xeon expandable processors enables real-time cardiac MRI applications, with data collected and immediately analyzed.
Cardiovascular disease accounts for a third of all deaths in the United States -- 34 per minute, or 18 million per year.Cardiac MRI has become the gold standard for evaluating cardiac function, ventricular volume, and cardiac tissue.
Cardiologists typically use manual or semi-automatic tools to extract quantitative measurements from cardiac magnetic resonance imaging (CMR), but this process is time-consuming, error-prone, and subject to subjectivity when interpreting images.
Dorin Comaniciu, senior vice President of Siemens healthcare, said: "based on Intel xlpi, we are now able to develop multiple real-time and utility-critical medical imaging use cases, such as cardiac MRI, without additional cost and complex hardware accelerators."
Heart models using artificial intelligence will save cardiologists more time by eliminating the need to manually segment images of the ventricles, heart muscle and blood vessels.When slice scanner to generate images, instantly in the edge of image segmentation based on artificial intelligence technology, make the deployed on the edge of the computing system can real-time capture the data, which brought artificial intelligence reasoning with low latency and high throughput speed advantage, such as medical institutions to serve more patients diagnosis safely every day.
The life sciences and health industry is transforming healthcare digitization by using artificial intelligence to accelerate clinical workflow, improve accuracy and diagnostics, and provide greater support for medical research while reducing hospital costs.Ai can quickly provide visualization of anatomical systems and identify abnormal conditions, which helps clinicians further focus on patient care.
Most of the systems Siemens healthcare deploys today use Intel processors, enabling Siemens healthcare to run ai reasoning workloads using existing cpu-based infrastructures.Siemens healthcare and Intel use the Intel DistribuTIon of OpenVINO toolkit to optimize, quantify, and execute models.The final demonstration results show that the speed has been improved by more than 5 times, and the accuracy has almost no loss.
Intel deep learning acceleration technology is a new set of embedded processor technology, which can accelerate the implementation of deep learning use cases.It extends the new vector neural network instruction (VNNI) in the Intel avx-512 instruction set,
This instruction is built into the second generation of Intel xeon expandable processors.Tasks such as convolution, which used to require three instructions, now require only one.The technology can be applied to target workloads including image recognition, image segmentation, speech recognition, language translation and object detection.