Deep learning-based arterial subtraction images improve the detection of LR-TR algorithm for viable HCC on extracellular agents-enhanced MRI

PURPOSE

To determine the role of deep learning-based arterial subtraction images in viability assessment on extracellular agents-enhanced MRI using LR-TR algorithm.

METHODS

Patients diagnosed with HCC who underwent locoregional therapy were retrospectively collected. We constructed a deep learning-based subtraction model and automatically generated arterial subtraction images. Two radiologists evaluated LR-TR category on ordinary images and then evaluated again on ordinary images plus arterial subtraction images after a 2-month washout period. The reference standard for viability was tumor stain on the digital subtraction hepatic angiography within 1 month after MRI.

RESULTS

286 observations of 105 patients were ultimately enrolled. 157 observations were viable and 129 observations were nonviable according to the reference standard. The sensitivity and accuracy of LR-TR algorithm for detecting viable HCC significantly increased with the application of arterial subtraction images (87.9% vs. 67.5%, p < 0.001; 86.4% vs. 75.9%, p < 0.001). And the specificity slightly decreased without significant difference when the arterial subtraction images were added (84.5% vs. 86.0%, p = 0.687). The AUC of LR-TR algorithm significantly increased with the addition of arterial subtraction images (0.862 vs. 0.768, p < 0.001). The arterial subtraction images also improved inter-reader agreement (0.857 vs. 0.727).

CONCLUSION

Extended application of deep learning-based arterial subtraction images on extracellular agents-enhanced MRI can increase the sensitivity of LR-TR algorithm for detecting viable HCC without significant change in specificity.

Evaluation of washout using subtraction MRI for the diagnosis of hepatocellular carcinoma in cirrhotic patients with spontaneously T1-hyperintense nodules

PURPOSE

The purpose of this study was to assess the value of subtraction imaging on post-arterial phase images (i.e., portal venous, delayed/transitional and hepatobiliary phases) for the non-invasive diagnosis of hepatocellular carcinoma (HCC) in spontaneously hyperintense nodules on T1-weighted imaging in patients with cirrhosis.

MATERIALS AND METHODS

Forty-five patients with a total 55 hepatic nodules that were spontaneously hyperintense on T1-weighted images were initially retrieved. All patients underwent MRI examination of the liver using extracellular agent. Each nodule was assessed for sensitivity and specificity using LI-RADS (Liver Imaging Reporting and Data System) during two reading sessions performed first without then with subtraction images on post-arterial phase images. The final standard of reference was defined by a step-by-step algorithm previously published combining histology, typical imaging, alfa fetoprotein and follow-up.

RESULTS

Forty-six nodules (26 HCC) in 39 patients with cirrhosis were analyzed. Using LI-RADS, the sensitivity and specificity for the diagnosis of HCC were 64% (95% CI: 41-83) and 67% (95% CI: 41-87) without subtraction; and 73% (95% CI: 50-89) (P > 0.999) and 33% (95% CI: 13-59) (P = 0.553) on subtraction imaging using extracellular contrast agent. Fifty-five percent (22/40) of nodules displayed a washout without subtraction and 70% (28/40) did so on subtraction imaging obtained with extracellular contrast agent. Twenty nodules out of 40 (50%) were classified LI-RADS 5 without subtraction, and 28 out of 40 nodules (70%) with subtraction.

CONCLUSION

The results of this study suggest that the use of subtraction imaging on post-arterial phase images (i.e., PVP, DP/TP and HBP) is not relevant for the non-invasive diagnosis of HCC for spontaneously hyperintense nodules on T1-weighted images in patients with liver cirrhosis.

The role of enhancing capsule and modified capsule appearances in LI-RADS for diagnosing HCC ≤ 3.0 cm on gadoxetate disodium-enhanced MRI

OBJECTIVES

To evaluate the value of using enhancing capsule (EC) or modified capsule appearance as a major feature in LI-RADS for diagnosing HCC ≤ 3.0 cm on gadoxetate disodium-enhanced MRI (Gd-EOB-MRI), and to explore the relationship between the imaging features and the histological fibrous capsule.

METHODS

This retrospective study enrolled 342 hepatic lesions ≤ 3.0 cm in 319 patients that underwent Gd-EOB-MRIs from January 2018 to March 2021. During dynamic phases and hepatobiliary phase, the modified capsule appearance added the nonenhancing capsule (NEC) (modified LI-RADS + NEC) or corona enhancement (CoE) (modified LI-RADS + CoE) to EC as an alternative capsule appearance. Inter-reader agreement of imaging features was assessed. The diagnostic performances of LI-RADS, LI-RADS with EC ignored, and two modified LI-RADS were compared, followed by Bonferroni correction. Multivariable regression analysis was performed to identify the independent features associated with the histological fibrous capsule.

RESULTS

The inter-reader agreement on EC (0.64) was lower than that on the NEC alternative (0.71) but better than that on CoE alternative (0.58). For HCC diagnosis, compared to LI-RADS, LI-RADS with EC ignored showed significantly lower sensitivity (72.7% vs. 67.4%, p < 0.001) with comparable specificity (89.3% vs. 90.7%, p = 1.000). Two modified LI-RADS showed slightly higher sensitivity and lower specificity than LI-RADS, without statistical significance (all p ≥ 0.006). The AUC was highest with modified LI-RADS + NEC (0.82). Both EC and NEC were significantly associated with the fibrous capsule (p < 0.05).

CONCLUSION

EC appearance improved the diagnostic sensitivity of LI-RADS for HCC ≤ 3.0 cm on Gd-EOB-MRI. Considering NEC as an alternative capsule appearance allowed for better inter-reader reliability and comparable diagnostic ability.

KEY POINTS

• Using the enhancing capsule as a major feature in LI-RADS significantly improved the sensitivity of diagnosing HCC ≤ 3.0 cm without reducing specificity on gadoxetate disodium-enhanced MRI. • Compared to the corona enhancement, the nonenhancing capsule might be a preferable alternative capsule appearance for diagnosing HCC ≤ 3.0 cm. • Capsule appearance should be considered a major feature in LI-RADS for diagnosing HCC ≤ 3.0 cm, regardless whether the capsule appears to be enhancing or nonenhancing.

A Multicenter Assessment of Interreader Reliability of LI-RADS Version 2018 for MRI and CT

Background Various limitations have impacted research evaluating reader agreement for Liver Imaging Reporting and Data System (LI-RADS). Purpose To assess reader agreement of LI-RADS in an international multicenter multireader setting using scrollable images. Materials and Methods This retrospective study used deidentified clinical multiphase CT and MRI and reports with at least one untreated observation from six institutions and three countries; only qualifying examinations were submitted. Examination dates were October 2017 to August 2018 at the coordinating center. One untreated observation per examination was randomly selected using observation identifiers, and its clinically assigned features were extracted from the report. The corresponding LI-RADS version 2018 category was computed as a rescored clinical read. Each examination was randomly assigned to two of 43 research readers who independently scored the observation. Agreement for an ordinal modified four-category LI-RADS scale (LR-1, definitely benign; LR-2, probably benign; LR-3, intermediate probability of malignancy; LR-4, probably hepatocellular carcinoma [HCC]; LR-5, definitely HCC; LR-M, probably malignant but not HCC specific; and LR-TIV, tumor in vein) was computed using intraclass correlation coefficients (ICCs). Agreement was also computed for dichotomized malignancy (LR-4, LR-5, LR-M, and LR-TIV), LR-5, and LR-M. Agreement was compared between research-versus-research reads and research-versus-clinical reads. Results The study population consisted of 484 patients (mean age, 62 years ± 10 [SD]; 156 women; 93 CT examinations, 391 MRI examinations). ICCs for ordinal LI-RADS, dichotomized malignancy, LR-5, and LR-M were 0.68 (95% CI: 0.61, 0.73), 0.63 (95% CI: 0.55, 0.70), 0.58 (95% CI: 0.50, 0.66), and 0.46 (95% CI: 0.31, 0.61) respectively. Research-versus-research reader agreement was higher than research-versus-clinical agreement for modified four-category LI-RADS (ICC, 0.68 vs 0.62, respectively; P = .03) and for dichotomized malignancy (ICC, 0.63 vs 0.53, respectively; P = .005), but not for LR-5 (P = .14) or LR-M (P = .94). Conclusion There was moderate agreement for LI-RADS version 2018 overall. For some comparisons, research-versus-research reader agreement was higher than research-versus-clinical reader agreement, indicating differences between the clinical and research environments that warrant further study. © RSNA, 2023 Supplemental material is available for this article. See also the editorials by Johnson and Galgano and Smith in this issue.

Optional MRI sequences for LI-RADS: why, what, and how?

Hepatocellular carcinoma (HCC) is the most common primary malignant tumor of the liver worldwide. Noninvasive diagnosis of HCC is possible based on imaging features, without the need for tissue diagnosis. Liver Imaging Reporting and Data System (LI-RADS) CT/MRI diagnostic algorithm allows for standardized radiological interpretation and reporting of imaging studies for patients at high risk for HCC. Diagnostic categories of LR-1 to LR-5 designate each liver observation to reflect the probability of overall malignancy, HCC, or benignity based on imaging features, where LR-5 category has > 95% probability of HCC. Optimal imaging protocol and scanning technique as described by the technical recommendations for LI-RADS are essential for the depiction of features to accurately characterize liver observations. The LI-RADS MRI technical guidelines recommend the minimum required sequences of T1-weighted out-of-phase and in-phase Imaging, T2-weighted Imaging, and multiphase T1-weighted Imaging. Additional sequences, including diffusion-weighted imaging, subtraction imaging, and the hepatobiliary phase when using gadobenate dimeglumine as contrast, improve diagnostic confidence, but are not required by the guidelines. These optional sequences can help differentiate true lesions from pseudolesions, detect additional observations, identify parenchymal observations when other sequences are suboptimal, and improve observations conspicuity. This manuscript reviews the optional sequences, the advantages they offer, and discusses technical optimization of these sequences to obtain the highest image quality and to avoid common artifacts.

Imaging Diagnosis of HCC: Future directions with special emphasis on hepatobiliary MRI and contrast-enhanced ultrasound

Hepatocellular carcinoma (HCC) is a unique cancer entity that can be noninvasively diagnosed using imaging modalities without pathologic confirmation. In 2018, several major guidelines for HCC were updated to include hepatobiliary contrast agent magnetic resonance imaging (HBA-MRI) and contrast-enhanced ultrasound (CEUS) as major imaging modalities for HCC diagnosis. HBA-MRI enables the achievement of high sensitivity in HCC detection using the hepatobiliary phase (HBP). CEUS is another imaging modality with real-time imaging capability, and it is reported to be useful as a second-line modality to increase sensitivity without losing specificity for HCC diagnosis. However, until now, there is an unsolved discrepancy among guidelines on whether to accept “HBP hypointensity” as a definite diagnostic criterion for HCC or include CEUS in the diagnostic algorithm for HCC diagnosis. Furthermore, there is variability in terminology and inconsistencies in the definition of imaging findings among guidelines; therefore, there is an unmet need for the development of a standardized lexicon. In this article, we review the performance and limitations of HBA-MRI and CEUS after guideline updates in 2018 and briefly introduce some future aspects of imaging-based HCC diagnosis.

Corona enhancement can substitute enhancing capsule in the imaging diagnosis of small (≤ 3 cm) HCCs on gadoxetic acid-enhanced MRI

OBJECTIVES

This study aimed to elucidate the relationship between gadoxetic acid-enhanced magnetic resonance imaging (MRI) features-enhancing capsule, corona enhancement or hypointense rim-observed in hepatocellular carcinomas (HCCs).

METHODS

Of the HCCs surgically confirmed during a 5-year period (2013-2017), ≤ 3-cm lesions (n = 83) in 78 patients were evaluated. Presence of corona enhancement and enhancing capsule on multiphasic dynamic imaging and presence of hypointense rim on hepatobiliary phase imaging were determined retrospectively by two independent observers. The relationship among the three imaging features was statistically analysed and correlated with the presence of histologic fibrous capsules, tumour differentiation and gross morphologic type.

RESULTS

There was substantial overall interobserver agreement in determining the presence of the three imaging features. Sixty (72.3%) lesions had histologic fibrous capsule positively correlated with all three imaging features (p < 0.05). Corona enhancement was the most common (66.3%) feature followed by enhancing capsule (61.4%) and hypointense rim (33.7%), and the correspondence rate of enhancing capsule to corona enhancement was 68.6% (p = 0.004). Corona enhancement was more frequently observed in moderately differentiated HCCs than other lesions (p = 0.012) and not dependent (p = 0.465) on the tumour size, while enhancing capsule was significantly dependent on tumour size, as indicated by univariate (p < 0.001) and multivariate analyses (odds ratio, 4.241; p = 0.002).

CONCLUSIONS

Among the capsular features, corona enhancement might closely relate to enhancing capsule in HCCs. Corona enhancement was not dependent on tumour size and had the highest incidence of appearance on gadoxetic acid-enhanced multiphasic dynamic MRI.

KEY POINTS

• Enhancing capsule has a limited role in the LI-RADS categorisation during gadoxetic acid-enhanced MRI. • Appearance of corona enhancement is closely related to enhancing capsule and is not dependent on size of HCCs. • Corona enhancement can substitute enhancing capsule in the diagnosis of HCCs during multiple arterial and portal venous phase gadoxetic acid-enhanced MRI.

The role of lesion hypointensity on gadobenate dimeglumine-enhanced hepatobiliary phase MRI as an additional major imaging feature for HCC classification using LI-RADS v2018 criteria

OBJECTIVES

To determine the value of lesion hypointensity in the hepatobiliary phase (HBP) on gadobenate dimeglumine-enhanced MRI as an additional major imaging feature for diagnosis of hepatocellular carcinoma (HCC) using LI-RADS v2018 criteria.

METHODS

Between March 2016 and August 2018, 235 patients with 250 hepatic nodules at high risk of HCC underwent gadobenate dimeglumine-enhanced MRI. Two radiologists independently evaluated the imaging features and classified the nodules based on LI-RADS v2018 criteria, and their consensus data were used to calculate the diagnostic performance of LI-RADS categories. Two modified LI-RADS definitions were as follows: (1) LI-RADS-m1: HBP hypointensity as an additional major feature; (2) LI-RADS-m2: HBP hypointensity as an alternative to "enhancing capsule" as an additional major feature. The diagnostic performance of LR-5 categories was compared using McNemar's test.

RESULTS

The sensitivity and specificity for LR-5 classification using original LI-RADS v2018 criteria were 78.1% and 96.3%, respectively. Significantly improved sensitivity (82.7%; p = 0.004) with unchanged specificity (96.3%; p = 1.00) was seen for LR-5 classification using LI-RADS-m1. Similar sensitivity and specificity (82.7% and 96.3%, respectively) were also seen using LI-RADS-m2. Significantly improved sensitivity (79.5% vs. 64.0%; p = 0.031) with unchanged specificity (96.2% vs. 96.2%, p = 1.00) was seen using both LI-RADS-m1 and LI-RADS-m2 compared to the original LI-RADS v2018 for 39 HCC nodules measuring 10-19 mm.

CONCLUSIONS

Lesion hypointensity on gadobenate dimeglumine-enhanced HBP MRI may improve sensitivity for LR-5 classification beyond that achievable using conventional LI-RADS v2018 criteria. Lesion hypointensity may prove a suitable alternative imaging feature to enhancing capsule for accurate LR-5 classification.

KEY POINTS

• Including lesion hypointensity in the HBP as an additional major feature improved sensitivity for LR-5 classification on gadobenate dimeglumine-enhanced MRI. • Lesion hypointensity in the HBP can replace "enhancing capsule" as an additional major feature for LR-5 classification without impairing specificity.