Partnership Tests Algorithms for Automatic Aorta Sizing
Emory Radiology enjoys a strong collaborative partnership with Siemens Healthineers, one that transcends the typical vendor-customer relationship. The two organizations collaborate on research that advances clinical applications of imaging technology.
For example, Carlo De Cecco, MD, PhD, associate professor, and Marly van Assen, PhD, post-doctoral researcher in Dr. De Cecco`s Lab, worked with Siemens to test artificial intelligence (AI) algorithms for automatic evaluation of the chest including aortic size, lung nodules, emphysema quantification, coronary artery calcium, vertebral fracture, and coronary arteries stenosis using computed tomography (CT). Such an algorithm will make it possible to evaluate chest CT and coronary CT angiography studies more accurately and quickly than with traditional methods, freeing radiologists to devote the bulk of their time to formulating treatment plans.
“It’s important to see how we can incorporate AI into the clinical setting,” explains Dr. De Cecco. “It’s not easy to do but getting it right will result in a huge advance for clinical care, one that saves time—which saves lives—and is cost-effective, which benefits more patients.”
The results are described in an article titled “Convolutional Neural Network Algorithm for Automatic Thoracic Aorta Sizing: Performance Testing in a Heterogeneous Population with Multivendor CT Datasets" submitted for publication.
For example, Carlo De Cecco, MD, PhD, associate professor, and Marly van Assen, PhD, post-doctoral researcher in Dr. De Cecco`s Lab, worked with Siemens to test artificial intelligence (AI) algorithms for automatic evaluation of the chest including aortic size, lung nodules, emphysema quantification, coronary artery calcium, vertebral fracture, and coronary arteries stenosis using computed tomography (CT). Such an algorithm will make it possible to evaluate chest CT and coronary CT angiography studies more accurately and quickly than with traditional methods, freeing radiologists to devote the bulk of their time to formulating treatment plans.
“It’s important to see how we can incorporate AI into the clinical setting,” explains Dr. De Cecco. “It’s not easy to do but getting it right will result in a huge advance for clinical care, one that saves time—which saves lives—and is cost-effective, which benefits more patients.”
The results are described in an article titled “Convolutional Neural Network Algorithm for Automatic Thoracic Aorta Sizing: Performance Testing in a Heterogeneous Population with Multivendor CT Datasets" submitted for publication.