AJR. American journal of roentgenology
-
AJR Am J Roentgenol · Apr 2017
Exaggerated Interventricular Dependence Among Patients With Pectus Excavatum: Combined Assessment With Cardiac MRI and Chest CT.
We sought to explore whether patients with pectus excavatum have exaggerated interventricular dependence and to evaluate the impact of the malformation severity (assessed on CT) on both anatomic and functional cardiac parameters (assessed on cardiac MRI). ⋯ In this study, patients with pectus excavatum showed significant alterations of cardiac morphology and function that were related to the deformation severity and that manifest as an exaggerated interventricular dependence.
-
AJR Am J Roentgenol · Apr 2017
Preoperative Breast MRI: Surgeons' Patient Selection Patterns and Potential Bias in Outcomes Analyses.
The purpose of this study is to determine which patient- and tumor-related and clinical variables influence dedicated breast surgeons' and general surgeons' referrals for preoperative breast MRI for patients with newly diagnosed breast cancer. ⋯ Preoperative breast MRI referral trended with certain higher risk patient- and tumor-related and clinical variables and were nonuniform between the breast surgeons and general surgeon cohorts. Selection bias could affect outcomes analyses for preoperative breast MRI.
-
AJR Am J Roentgenol · Apr 2017
Multicenter StudyPerformance of a Machine Learning Classifier of Knee MRI Reports in Two Large Academic Radiology Practices: A Tool to Estimate Diagnostic Yield.
The purpose of this study is to evaluate the performance of a natural language processing (NLP) system in classifying a database of free-text knee MRI reports at two separate academic radiology practices. ⋯ The results show excellent accuracy by the NLP machine learning classifier in classifying free-text knee MRI reports, supporting the institution-independent reproducibility of knee MRI report classification. Furthermore, the machine learning classifier performed well on free-text knee MRI reports from another institution. These data support the feasibility of multiinstitutional classification of radiologic imaging text reports with a single machine learning classifier without requiring institution-specific training data.