CAD Algorithms for Solid Breast Masses Discrimination: Evaluation of the Accuracy and Interobserver Variability

Evaluation of the Accuracy of a Computer-aided Diagnosis (CAD) System in Breast Ultrasound according to the Radiologist’s Experience

“…However, the studies available on this subject showed an improvement of the radiologists' sensitivity scores, including those of their less experienced counterparts. Wang et al (26) evaluated the impact of CAD on eight radiologists with different levels of expertise (four experienced radiologists and four residents). The sensitivity scores of senior radiologists (90.1% vs 85.24%) and especially resident radiologists (89.46% vs 80.4%) were improved when interpretation was done using CAD.…”

“…The results were similar in Sahiner et al's study (22), with three-dimensional ultrasound CAD, and their study showed even a slight decrease in the specificity scores of expert radiologists (19% vs 22%, P = NS) with CAD. In the study by Wang et al (26), which assessed the impact of CAD on 8 different levels of experience among radiologists (four experienced radiologists and four residents), the specificity score was improved during image interpretation with CAD, for both senior radiologists (from 68.2% to 74.4%) and residents (from 62.6% to 65%). Differences were found in the sensitivity and specificity scores of the same radiologist between benign and malignant lesions and classification according to BI-RADS categories.…”

Breast Ultrasound Computer-Aided Diagnosis: Analysis of Types of Errors

“…One is the interpretation according to the breast imaging reporting and data system (BI-RADS) lexicon, which is known as the knowledge-based analysis, and the other is the deep-learning algorithm, which is known as statistics-based analysis. that is based on a deep-learning algorithm that uses big data and provides assistance in morphological analysis based on the BI-RADS lexicon and final assessment (4,12). S-Detect might be very useful in improving the diagnostic performance of breast US for assistance in either lesion detection or the decisionmaking process during practice.…”

Computer‐aided diagnosis program for classifying the risk of hepatocellular carcinoma on MR images following liver imaging reporting and data system (LI‐RADS)

“…Computer‐aided diagnosis (CAD) is a potential solution to reduce inter‐ or intrareader variability, increase efficiency, and reduce diagnostic error . Studies in breast imaging showed that computer‐aided analysis of BI‐RADS features provided more precise approaches to characterize breast lesions, improve diagnostic accuracy, and reduce interobserver variability . Compared to breast imaging, the application of CAD systems for the characterization of liver lesions is less well established .…”

“…8 Studies in breast imaging showed that computer-aided analysis of BI-RADS features provided more precise approaches to characterize breast lesions, [9][10][11][12][13] improve diagnostic accuracy, 14,15 and reduce interobserver variability. 16 Compared to breast imaging, the application of CAD systems for the characterization of liver lesions is less well established. 17,18 In particular, no study has been reported for the development of CAD for LI-RADS.…”