Focal Liver Lesions: Computer-aided Diagnosis by Using Contrast-enhanced US Cine Recordings

Abstract

Purpose To assess the performance of computer-aided diagnosis (CAD) systems and to determine the dominant ultrasonographic (US) features when classifying benign versus malignant focal liver lesions (FLLs) by using contrast material-enhanced US cine clips. Materials and Methods One hundred six US data sets in all subjects enrolled by three centers from a multicenter trial that included 54 malignant, 51 benign, and one indeterminate FLL were retrospectively analyzed. The 105 benign or malignant lesions were confirmed at histologic examination, contrast-enhanced computed tomography (CT), dynamic contrast-enhanced magnetic resonance (MR) imaging, and/or 6 or more months of clinical follow-up. Data sets included 3-minute cine clips that were automatically corrected for in-plane motion and automatically filtered out frames acquired off plane. B-mode and contrast-specific features were automatically extracted on a pixel-by-pixel basis and analyzed by using an artificial neural network (ANN) and a support vector machine (SVM). Areas under the receiver operating characteristic curve (AUCs) for CAD were compared with those for one experienced and one inexperienced blinded reader. A third observer graded cine quality to assess its effects on CAD performance. Results CAD, the inexperienced observer, and the experienced observer were able to analyze 95, 100, and 102 cine clips, respectively. The AUCs for the SVM, ANN, and experienced and inexperienced observers were 0.883 (95% confidence interval [CI]: 0.793, 0.940), 0.829 (95% CI: 0.724, 0.901), 0.843 (95% CI: 0.756, 0.903), and 0.702 (95% CI: 0.586, 0.782), respectively; only the difference between SVM and the inexperienced observer was statistically significant. Accuracy improved from 71.3% (67 of 94; 95% CI: 60.6%, 79.8%) to 87.7% (57 of 65; 95% CI: 78.5%, 93.8%) and from 80.9% (76 of 94; 95% CI: 72.3%, 88.3%) to 90.3% (65 of 72; 95% CI: 80.6%, 95.8%) when CAD was in agreement with the inexperienced reader and when it was in agreement with the experienced reader, respectively. B-mode heterogeneity and contrast material washout were the most discriminating features selected by CAD for all iterations. CAD selected time-based time-intensity curve (TIC) features 99.0% (207 of 209) of the time to classify FLLs, versus 1.0% (two of 209) of the time for intensity-based features. None of the 15 video-quality criteria had a statistically significant effect on CAD accuracy-all P values were greater than the Holm-Sidak α-level correction for multiple comparisons. Conclusion CAD systems classified benign and malignant FLLs with an accuracy similar to that of an expert reader. CAD improved the accuracy of both readers. Time-based features of TIC were more discriminating than intensity-based features. RSNA, 2017 Online supplemental material is available for this article.

Differentiation of atypical hepatic hemangioma from liver metastases: Diagnostic performance of a novel type of color contrast enhanced ultrasound

“…Ultrasonography is useful for the diagnosis of hepatic focal lesions, which is based on their distinctive echogenicities-the grayscale morphologic features. The development of ultrasound contrast agents provided us with more information about tissue perfusion and helped to improve diagnostic accuracy, particularly in focal liver lesions (FLLs) [15][16][17] . Compared with first-generation agents, the second-generation ultrasound contrast agent SonoVue consisting of sulfur hexafluoride microbubbles (Bracco, Milan, Italy) has a high flexibility shell and is more stable to acoustic pressure.…”

Artificial intelligence assists identifying malignant versus benign liver lesions using contrast‐enhanced ultrasound

“… 30 , 31 For AI‐US, our study reported the largest sample size with an independent test dataset. 32 , 33 , 34 Compared with a previous AI‐US study with an independent test dataset, 35 our AI model exhibited better performance (AUC: 93.4% vs 88.1%).…”

“…Its performance was comparable with or even better than the previously reported performance of non‐AI CT (Se: 89%, Sp: 94%) and MR (Se: 83%, Sp: 75%). 7 A CEUS‐based AI model was also reported in a recent study of FLL differentiation 34 ; however, that study used machine learning algorithms based on manually extracted features for model development. That study reported an Se of 83.3% and an Sp of 62.7%, and these values are lower than those obtained with our deep learning model.…”

Is There a Place for Elastography in the Diagnosis of Hepatocellular Carcinoma?

“…Several proposed practical algorithms for the clinical use of MPUS in chronic liver disease and FLL are available [ 24 ], and computer-aided diagnosis systems were conceived based on the new US features [ 25 , 26 ].…”