Medical AI Startup Quickly Shifts Strategy to Fight COVID-19 Pandemic

China’s Shukun Technology Uses NVIDIA GPUs to Develop Lung Doc, an AI Tool That Speeds Analysis COVID-19 of Chest CT Scans.

The COVID-19 pandemic has disrupted the world like few events before it.

But for Shukun Technology, a response required “a minor change in our strategy,” according to its chief technology officer, Chao Zheng.

That’s because Shukun, a startup founded by some of China’s brightest AI and medical minds, was busy refining its AI-powered platform to diagnose heart disease and strokes when the global pandemic struck.

The company quickly shifted resources to develop a system that analyzes chest CT scans to help speed up diagnoses of COVID-19 patients. That system, called Lung Doc – pneumonia edition, has already been rolled out to 30 hospitals in China over the past few months, where it will grow more accurate as it learns from more data.

Powered by NVIDIA GPUs, the system is proving so effective, according to Zheng, that Shukun is working on beefing it up for use in the many other countries that have inquired about using it.

Recommended AI News: Sigma Computing Makes the Data-Driven Marketing Dream a Reality with New Analytics and Business Intelligence Templates

Super-Charging Radiologists

Prior to Lung Doc, the three-year-old company had introduced two suites of products focused on heart ailments and strokes. At its essence, Shukun’s technology serves a very specific and valuable function: Shortening the time radiologists spend reconstructing and analyzing 2D and 3D images, and making diagnoses based on their findings.

Typically, these tasks require about a half-hour of a skilled radiologist’s time. But radiologists use varying techniques and approaches, so some take longer. Shukun’s AI adds consistency and speed, requiring just a minute to perform both the reconstruction and diagnosis processes.

“It’s a very efficient tool that helps radiologists deliver quicker, more accurate results,” said Zheng.

Little public data is available for life-threatening diseases, so in building its AI models, Shukun has turned to a vast network of hospital, academic and research partners to obtain private datasets totaling more than 100,000 cases, each of which typically contains 200-300 images.

As the company has worked on its models, it’s relied on blending imaging data to support segmentation and classification work. Transfer learning has helped speed the development process by applying lessons from one disease across others.

GPUs Delivering Results

NVIDIA GPUs have played a critical role for Shukun, with a combination of NVIDIA V100 Tensor Core and P100 GPUs (more than 500 in all) used for training, while NVIDIA T4 data center GPUs handle inference locally. The company can set up hospitals to run the system completely on premises or in a private cloud-like environment.

Zheng said each of these GPUs is delivering 20 times the performance of previous generations of GPUs and CPUs, and heart disease and stroke products have achieved accuracy rates that rival human detection abilities.

While future GPUs are expected to continue speeding up the training process, and thus ultimately the accuracy, Zheng said that a federated learning model, with hospitals and Shukun’s data centers all exchanging data in real time, would help the company’s cause.

For now, Shukun has to work with each hospital individually to obtain fresh data, and connectivity at individual facilities is often an issue. If each hospital could be connected to a federated data model, only local training would be required for them to access all the data flowing through Shukun’s network of partners.

Recommended AI News: Core Scientific Introduces The Cloud for Data Scientists at Equinix – Powered by NVIDIA and NetApp

In Pursuit of Federated Data

Shukun’s heart and stroke products are deployed in more than 200 hospitals throughout China. Many of these hospitals have already fully integrated the company’s AI into their radiology and imaging departments, while some are choosing to invest a month or two in clinical trials before using the technology with live patient data.

As Shukun adds more data and builds its federated model, it plans to expand its horizons to tackle more diseases and modalities. For instance, it would like to develop MRI solutions, which would then open the door to introducing its AI in more clinical settings.

The company is also investigating the NVIDIA Clara application framework for its federated learning capabilities, which let researchers and data scientists collaborate on AI algorithms, while keeping data private and secure.

Said Zheng: “We plan to spread AI throughout the medical community.”

In the meantime, the company will continue doing its part to help the world move closer to controlling the COVID-19 pandemic.

Recommended AI News: Bitcoin Association Publishes First Annual Report Highlighting Rapid Growth Of Bitcoin SV Ecosystem