The machine learning-based method can ensure clinicians keep pace with the growing number of people presenting with more than one lung abnormality.
The deep learning method was trained on more than 800,000 echocardiogram videos gathered over the past decade, researchers explained Monday.
The American College of Radiology's catalog includes 111 class 2 medical imaging algorithms searchable by company, modality, subspeciality, and more.
Under a new proposal, 91 products would no longer require U.S. Food & Drug Administration regulatory review, including a number of radiology-driven platforms.
The artificial intelligence platform outperformed both body mass index and overall weight at forecasting such outcomes, researchers explained recently.
During a session at RSNA, researchers detailed how their 2D convolutional neural network outperformed even the best-trained algorithms.
The 10 teams will share $30,000 in prize money and were recognized during the society's annual meeting.
Some are calling on the agency to boost proposed payments for a tool that diagnoses diabetic retinopathy in order to encourage widespread adoption.
Actions taken at Changi General Hospital in Singapore led to a 17.2% improvement from the institution’s baseline missed appointment rate.
In total, 7,774 images taken from 287 patients were used to train the deep learning model, according to a study published in AJR.
H. Lee Moffitt Cancer Center and Research Institute scientists used low-dose CT and chest x-ray imaging data from the National Lung Screening Trial to create their model.
The technique—known as 3D cartilage surface mapping—detects subtle changes in a person’s knee joint that cannot be picked up via conventional x-ray or MRI.