Using AI to identify COVID-19 lesions in lung CT scans
Mike McNamara
The high numbers of hospitalizations and the level of critical care that many COVID-19 patients require can push healthcare institutions and staff to their limits. COVID pneumonia (viral infection in the lungs), which is detected by a chest x-ray or CT scan, can predict the need for more advanced inpatient care.
A busy hospital may perform many lung CTs per day, potentially affecting the service levels that radiology teams are able to deliver. By prescreening the CT scans of COVID-19 patients, an accurate AI model can quickly pinpoint critical results and enable care teams to zero in on patients who are at high risk for severe complications.
Model tuning, testing, and ongoing training are necessary to create and sustain an optimized artificial intelligence model. Careful attention to traceability, reproducibility, and patient privacy are essential. NetApp and SFL Scientific have developed technology for high-performing COVID-19 lung segmentation that uses a state-of-the-art model and transfer learning. The following image compares human annotations and model prediction of lung lesions in a COVID-19 patient. Our methodology delivers an accurate, trained model in a short time and supports ongoing training and optimization with complete traceability.
Running on a fast and efficient NetApp® storage infrastructure, the model takes an average of just 6 seconds to identify the COVID lesions on each patient scan, which is composed of hundreds of images. This speed is on par with other advanced models and is much faster than a typical human analysis of a chest CT.
Additional AI opportunities
The methodology that NetApp and SFL Scientific used to create a COVID-19 lung segmentation model can be generalized and applied to almost any image segmentation task. With access to the appropriate data, we can help you create AI segmentation models for any organ system, encompassing a wide range of imaging modalities, from simple 2D x-rays to 3D CT and MRI scans to ultrasound. Similar methods can also be applied to digital pathology.Looking beyond medical imaging, the same approach—combining transfer learning, experimentation, iterative fine tuning, intelligent data management, and production deployment with regular retraining—can be applied to a wide range of computer vision, natural language processing, and other use cases in healthcare and other industries. NetApp and SFL Scientific can help you get your AI project to production more quickly with fewer missteps.
To learn more, read the white paper Deep learning to identify COVID-19 lesions in lung CT scans and watch the on-demand video COVID-19 lung CT lesion segmentation.
Mike McNamara
Mike McNamara is a senior product and solution marketing leader at NetApp with over 25 years of data management and cloud storage marketing experience. Before joining NetApp over ten years ago, Mike worked at Adaptec, Dell EMC, and HPE. Mike was a key team leader driving the launch of a first-party cloud storage offering and the industry’s first cloud-connected AI/ML solution (NetApp), unified scale-out and hybrid cloud storage system and software (NetApp), iSCSI and SAS storage system and software (Adaptec), and Fibre Channel storage system (EMC CLARiiON).
In addition to his past role as marketing chairperson for the Fibre Channel Industry Association, he is a member of the Ethernet Technology Summit Conference Advisory Board, a member of the Ethernet Alliance, a regular contributor to industry journals, and a frequent event speaker. Mike also published a book through FriesenPress titled "Scale-Out Storage - The Next Frontier in Enterprise Data Management" and was listed as a top 50 B2B product marketer to watch by Kapos.
