Israel’s top medical center partners with AI startups


Israel’s largest private medical center is working with startups and researchers to bring potentially life-saving AI solutions to real-world healthcare workflows.

With over 1.5 million patients in eight medical centers, Assuta Medical Centers perform over 100,000 surgeries, 800,000 imaging tests and hundreds of thousands of other health diagnoses and treatments each year. These create huge amounts of anonymized data that Assuta securely shares with over 20 startups through its innovation arm, ASCENDlaunched last year in collaboration with Nvidia.

One of the startups, Aidoc, is helping Assuta alert imaging technicians with AI-powered information about possible brain bleeds and other critical conditions in a patient’s scan within minutes. Another, Rhino Health, uses federated learning powered by NVIDIA FLARE to make AI development on diverse medical datasets from hospitals around the world more accessible to Assuta employees.

Both companies are members of Creation of NVIDIAa global program designed to support cutting-edge startups with go-to-market support, expertise and technology.

“We are building a hub to provide innovators with the infrastructure they need to develop, test and deploy new AI technology for image analysis and other data-intensive computations in radiology, pathology, genomics and more,” said Daniel Rabina, Chief Innovation Officer. at Rise. “We want to make it possible to collaborate with companies, research institutes, hospitals and universities while preserving the confidentiality of patient data.”

To support AI development, testing and deployment, Assuta has installed NVIDIA DGX A100 Systems on the spot and adopted the NVIDIA Clara Holoscan platform, as well as software libraries, including MONAI for medical imaging and NVIDIA FLARE for federated learning.

NVIDIA and RISE collaborate on RISE with US, a program designed to introduce selected Israeli entrepreneurs and early-stage startups working on digital and IT health solutions to the US market. Applications to join the program are open until August 28.

Aidoc flags urgent cases for examination by a radiologist

Aidocwhich is based in New York with a research branch in Israel, has developed FDA-cleared AI solutions to flag acute conditions, including brain hemorrhages, pulmonary embolisms and strokes triggered by brain scans. imagery.

Founded in 2016 by a group of Israel Defense Forces veterans, the startup has deployed its AI to analyze millions of cases at more than 1,000 medical institutions, mostly in the US, Europe and Israel.

Its algorithms integrate seamlessly with the PACS imaging workflow used by radiologists around the world, working behind the scenes to analyze each imaging study and flag urgent results, bringing potentially critical cases to attention. from the radiologist for examination.

Aidoc’s tools can help address growth worldwide shortage of radiologists reducing the time a radiologist must spend on each case, allowing more patients to be treated. And by pushing potentially critical cases to the top of a radiologist’s pile, AI can help clinicians detect important findings more quickly, improving patient outcomes.

The startup uses NVIDIA Tensor Core GPUs in the cloud through AWS for AI training and inference. Adopting NVIDIA GPUs Reduced Model Training Time from a few days to a few hours.

Immediate impact in Assuta medical centers

Assuta is a private chain of hospitals that provides elective care – typically treating routine screenings rather than ER patients – but has adopted Aidoc’s solution to help imaging technicians spot critical cases that need attention among its approximately 200,000 CT tests performed each year.

When an X-ray analysis is not urgent, a doctor may take a few days to review the case. Aidoc can reduce this time to minutes by identifying affected cases as soon as scans are captured by radiology staff. Assuta facilities

At Assuta, urgent findings are usually found in cancer patients or people who have recently had surgery and need follow-up examinations. Healthcare organization uses AIdoc’s AI tools to detect intracranial hemorrhage and two types of pulmonary embolism.

“We saw the impact right away,” said Dr Michal Guindy, Head of Medical Imaging and Head of RISE at Assuta. “Just a few days after the installation of Aidoc in Assuta, a patient came for follow-up examination after brain surgery and had an intracranial hemorrhage. Because Aidoc alerted the imaging technician to flag him for further examination, our doctors were able to call the patient as he was returning home and immediately redirect him to the hospital for treatment.

Rhino Health encourages collaboration with federated learning

In addition to deploying AI models in real-world environments at scale, Assuta supports innovators who develop, test or validate new medical AI solutions by sharing healthcare organization data, while using federated learning via Rhino health.

Assuta has digitized millions of radiology cases – a desirable resource for researchers and startups looking for robust and diverse datasets to train or validate their AI models. But due to data privacy protection, it is important that patient information stays safe within the firewall of medical centers like Assuta.

“Diversity of data is needed to develop AI models for use by medical teams. Without optimal computing resources, it would be extremely difficult to use our data and perform the magic,” Rabina said. “That’s why we need federated learning enabled by both NVIDIA and Rhino Health.”

Federated learning enables businesses, healthcare institutions, and universities to work together by training and validating AI models on multiple organizations’ datasets while maintaining the privacy of each organization’s data. Rhino Health provides a neutral platform – available through the NVIDIA AI Enterprise software suite – which enables secure collaboration, powered by NVIDIA A100 GPUs in the cloud and NVIDIA FLARE federated learning framework.

With Rhino Health, Assuta aims to help its collaborators develop AI models in hospitals around the world, which will result in more generalizable algorithms that work more precisely on different patient populations.

Register for NVIDIA GTConline September 19-22, to learn about AI leaders in healthcare.

To subscribe to NVIDIA Health News and watch on demand like Assuta, Aidoc and Rhino Health speak at a GTC panel.


Comments are closed.