Sign In

Communications of the ACM

ACM TechNews

AI Program Accurately Predicts Lung Cancer Risk

View as: Print Mobile App Share:
A lung nodule identified in a computed tomography scan.

An artificial intelligence program can accurately predict the risk of lung nodules detected in computed tomography scans becoming cancerous,

Credit: Radiological Society of North America

A deep learning artificial intelligence (AI) algorithm developed by a team of Dutch, Danish, and German researcher can accurately predict the risk of lung nodules detected on screening computed tomography (CT) images turning cancerous.

The researchers trained the algorithm on CT images of over 16,000 nodules, including 1,249 malignancies, and validated it on three nodule imaging datasets. The algorithm outperformed the established Pan-Canadian Early Detection of Lung Cancer model for lung nodule malignancy risk estimation, and exhibited comparable performance to 11 clinicians. Kiran Vaidhya Venkadesh at the Netherlands' Radboud University said, "The algorithm may aid radiologists in accurately estimating the malignancy risk of pulmonary nodules. This may help in optimizing follow-up recommendations for lung cancer screening participants."

From "AI Program Accurately Predicts Lung Cancer Risk"

SciTechDaily (05/18/21)
View Full Article


No entries found