Radiologists spend more time reviewing mammogram images from women who were called back for further investigation, and these images were interpreted with more confidence.
Radiologists spend more time reviewing mammogram images from women who were called back for further investigation than from those who were not, and they were more confident in their findings, according to researchers in the April issue of American Journal of Roentgenology.
The researchers, from Oregon Health and Science University, undertook the study to examine if the time spent evaluating images affected the accuracy and confidence levels of the radiologists. In the study, 119 radiologists from six mammography registries were randomized to interpret one of four test sets, after which they completed 12 survey questions. Each test set, screening exams of women aged 40 to 69 who did not have a history of breast cancer, had 109 cases of digitized four-view screening screen-film mammograms, with prior comparison screening views. In total, the radiologists contributed data on 11,484 interpretations.
Viewing time was defined as the cumulative time spent viewing the images before the radiologists recorded findings. The researchers found that the radiologists spent more time reviewing the images that had significant findings or images in which they were not as confident with their interpretations.
“Each additional minute of viewing time increased the probability of a true-positive interpretation among cancer cases by 1.12. regardless of confidence in the assessment. Among the radiologists who were very confident in their assessment, each additional minute of viewing time increased the adjusted risk of a false-positive interpretation among noncancer cases by 1.42, and this viewing time diminished with decreasing confidence,” the authors wrote. Further, each additional minute of viewing time increased the probability of a true-positive interpretation among cancer cases, the authors noted.
Study: AI Bolsters Sensitivity for Pneumothorax on CXR and Significantly Reduces Reporting Time
October 30th 2024For clinically actionable pneumothorax, an artificial intelligence algorithm demonstrated a 93 percent AUC and a 96 percent specificity rate in a study involving chest X-rays from over 27,000 adults.
AI Segmentation, Intraprostatic Tumor Volume and Metastases: What a New mpMRI Study Reveals
October 29th 2024For patients who received radiotherapy for localized prostate cancer, total intraprostatic tumor volume derived from AI segmentation had a 10 percent higher AUC for predicting seven-year metastases in comparison to a risk model from the National Comprehensive Cancer Network (NCCN).