Leesburg, VA, August 17, 2022--According to ARRS' American Journal of Roentgenology (AJR), machine learning models applied to presently underutilized imaging features could help construct more reliable criteria for organ allocation and liver transplant eligibility. "The findings suggest that machine learning-based models can predict recurrence before therapy allocation in patients with early-stage hepatocellular carcinoma (HCC) initially eligible for liver transplant," wrote corresponding author Julius Chapiro from the department of radiology and biomedical imaging at Yale University School of Medicine in New Haven, CT. Chapiro and colleagues' proof-of-concept study included 120 patients (88 men, 32 women; median age, 60 years) diagnosed with early-stage HCC between June 2005 and March 2018, who were initially eligible for liver transplant and underwent treatment by transplant, resection, or thermal ablation. Patients underwent pretreatment MRI and posttreatment imaging surveillance, and imaging features were extracted from postcontrast phases of pretreatment MRI examinations using a pretrained convolutional neural network (VGG-16). Pretreatment clinical characteristics (including laboratory data) and extracted imaging features were integrated to develop three ML models--clinical, imaging, combined--for recurrence prediction within 1–6 years posttreatment. Ultimately, all three models predicted posttreatment recurrence for early-stage HCC from pretreatment clinical (AUC 0.60–0.78,

Diagnosing emphysema and classifying its severity have long been more art than science. "Everybody has a different trigger threshold for what they would call normal and what they would call disease," said U. Joseph Schoepf, M.D., director of cardiovascular imaging for MUSC Health and assistant dean for clinical research in the Medical University of South Carolina College of Medicine. And until recently, scans of damaged lungs have been a moot point, he said. In the past, if you lost lung tissue, that was it. The lung tissue was gone, and there was very little you could do in terms of therapy to help patients.

All publication searches were performed using a comprehensive central database (Web of Science Core Collection, Clarivate Analytics) that searches the world's leading scholarly journals and proceedings in the sciences and includes the MEDLINE and PubMed databases. From 2000 to 2018, all AI-related publications were selected using the following search terms: "artificial intelligence," "AI," "CNN," "CNNs," "ANN," "ANNs," "neural network," "neural networks," "machine learning," "deep learning," "computer learning," "support vector machine," "support vector machines," "Bayesian network," "Bayesian networks," "cluster analysis," "feature learning," "feature extraction," and "principal components analysis." Radiology-specific AI research was selected using the predefined database category "Radiology Nuclear Medicine Medical Imaging." The resulting publication database was then categorized by country of origin, funding agencies, organizations, publication type, and journal. Nine radiology subspecialty publications were evaluated using the following search terms.

I thank Keith Dryer and Bradley Erickson for their expertise, leadership, and educational efforts in the applications of AI in radiology, all of which have helped me understand this complex subject more fully.