European radiology
-
To conduct a multireader validation study to evaluate the interobserver variability and the diagnostic accuracy for the lung involvement by COVID-19 of COVID-19 Reporting and Data System (CO-RADS) score. ⋯ • COVID-19 Reporting and Data System (CO-RADS) demonstrated a good diagnostic accuracy for lung involvement by COVID-19 with an average AUC of 0.72 (95% CI 67-75%). • When a threshold of ≥ 4 was used, sensitivity and specificity were 61% (95% CI 52-69%) and 81% (95% CI 76-84%), respectively. • There was an overall moderate agreement for CO-RADS rating across readers with different levels of expertise (Fleiss' K = 0.43 [95% CI 0.42-0.44]).
-
Differentiation of malignant and benign pancreatic lesions on anatomical imaging is difficult in some cases with overlapping features. Prostate-specific membrane antigen (PSMA) is overexpressed during angioneogenesis in many tumors. We aimed to evaluate the PSMA expression in pancreatic lesions to differentiate these lesions and explore the performance of Ga-68 PSMA-PET/CT vis-a-vis F-18 FDG-PET/CT. ⋯ • Conventional imaging such as CT and MRI are unable to reliably differentiate localized malignant pancreatic lesion from benign lesions mimicking malignancy such as mass-forming pancreatitis. • FDG PET/CT helps in detecting malignant foci in view of their increased glucose metabolism. However, it may be falsely positive in inflammatory lesions which may occasionally hinder its ability to differentiate between benign and malignant lesions. • Apart from prostatic malignancy, PSMA is overexpressed in neovasculature of many non-prostatic malignancies. The present study highlights that Ga68 PSMA PET/CT performed better in diagnosing malignancy non-invasively than FDG-PET/CT with a higher PPV (90.5% vs. 65.4%) and accuracy (92.5% vs. 72.5%).
-
Why is there a major gap between the promises of AI and its applications in the domain of diagnostic radiology? To answer this question, we systematically review and critically analyze the AI applications in the radiology domain. ⋯ • Many AI applications are introduced to the radiology domain and their number and diversity grow very fast. • Most of the AI applications are narrow in terms of modality, body part, and pathology. • A lot of applications focus on supporting "perception" and "reasoning" tasks.