Deep learning-based compressed SENSE improved diffusion-weighted image quality and liver cancer detection: A prospective study.

PURPOSE: To assess whether diffusion-weighted imaging (DWI) with Compressed SENSE (CS) and deep learning (DL-CS-DWI) can improve image quality and lesion detection in patients at risk for hepatocellular carcinoma (HCC). METHODS: This single-center prospective study enrolled consecutive at-risk participants who underwent 3.0 T gadoxetate disodium-enhanced MRI. Conventional DWI was acquired using parallel imaging (PI) with SENSE (PI-DWI). In CS-DWI and DL-CS-DWI, CS but not PI with SENSE was used to accelerate the scan with 2.5 as the acceleration factor. Qualitative and quantitative image quality were independently assessed by two masked reviewers, and were compared using the Wilcoxon signed-rank test. The detection rates of clinically-relevant (LR-4/5/M based on the Liver Imaging Reporting and Data System v2018) liver lesions for each DWI sequence were independently evaluated by another two masked reviewers against their consensus assessments based on all available non-DWI sequences, and were compared by the McNemar test. RESULTS: 67 participants (median age, 58.0 years; 56 males) with 197 clinically-relevant liver lesions were enrolled. Among the three DWI sequences, DL-CS-DWI showed the best qualitative and quantitative image qualities (p range, <0.001-0.039). For clinically-relevant liver lesions, the detection rates (91.4%-93.4%) of DL-CS-DWI showed no difference with CS-DWI (87.3%-89.8%, p = 0.230-0.231) but were superior to PI-DWI (82.7%-85.8%, p = 0.015-0.025). For lesions located in the hepatic dome, DL-CS-DWI demonstrated the highest detection rates (94.8%-97.4% vs 76.9%-79.5% vs 64.1%-69.2%, p = 0.002-0.045) among the three DWI sequences. CONCLUSION: In patients at high-risk for HCC, DL-CS-DWI improved image quality and detection for clinically-relevant liver lesions, especially for the hepatic dome.

Noninvasive diagnosis of liver cirrhosis: qualitative and quantitative imaging biomarkers.

A diagnosis of cirrhosis initiates a shift in the management of chronic liver disease and affects the diagnostic workflow and treatment decision of primary liver cancer. Liver biopsy remains the gold standard for cirrhosis diagnosis, but it is invasive and susceptible to sampling bias and observer variability. Various qualitative and quantitative imaging biomarkers based on ultrasound, CT and MRI have been proposed for noninvasive diagnosis of cirrhosis. Qualitative imaging features are easy to apply but have moderate diagnostic sensitivity. Elastography techniques allow quantitative assessment of liver stiffness and are highly accurate for cirrhosis diagnosis. Ultrasound elastography are widely used in clinical practice, while MR elastography has narrower availability. Although not applicable in clinical practice yet, other quantitative imaging features, including liver surface nodularity, linear and volumetric measurement, extracellular volume fraction, liver enhancement on hepatobiliary phase, and parameters derived from diffusion-weighted imaging, can provide additional information of liver morphology, perfusion, and function, thus may increase diagnosis performance. The introduction of radiomics and deep learning has further improved diagnostic accuracy while reducing subjectivity. Several imaging features may also help to assess liver function and outcomes in patients with cirrhosis. In this review, we summarize the qualitative and quantitative imaging biomarkers for noninvasive cirrhosis diagnosis, and the assessment of liver function and outcomes, and discuss the challenges and future directions in this field.

Prediction of late recurrence after curative-intent resection using MRI-measured spleen volume in patients with hepatocellular carcinoma and cirrhosis.

BACKGROUND: Late recurrence of hepatocellular carcinoma (HCC) after liver resection is regarded as a de novo tumor primarily related to the severity of underlying liver disease. We aimed to investigate risk factors, especially spleen volume, associated with late recurrence in patients with HCC and cirrhosis. METHODS: We retrospectively analyzed 301 patients with HCC and cirrhosis who received curative resection and preoperative MRI. Patients were followed for late recurrence for at least 2 years. Spleen volume was automatically measured on MRI with artificial intelligence techniques, and qualitative MRI imaging features reflecting tumor aggressiveness were evaluated. Uni- and multivariable Cox regression analyses were performed to identify independent predictors and a risk score was developed to predict late recurrence. RESULTS: Eighty-four (27.9%) patients developed late recurrence during follow-up. Preoperative spleen volume was independently associated with late recurrence, and patients with a volume > 370 cm3 had significantly higher recurrence risk (hazard ratio 2.02, 95%CI 1.31-3.12, p = 0.002). Meanwhile, no qualitative imaging features were associated with late recurrence. A risk score was developed based on the APRI score, spleen volume, and tumor number, which had time-dependent area under the curve ranging from 0.700 to 0.751. The risk score at a cutoff of 0.42 allowed for the identification of two risk categories with distinct risk of late recurrence. CONCLUSIONS: Preoperative spleen volume on MRI was independently associated with late recurrence after curative-intent resection in patients with HCC and cirrhosis. A risk score was proposed for individualized risk prediction and tailoring of postoperative surveillance strategies. CRITICAL RELEVANCE STATEMENT: Spleen volume measured on MRI with the aid of AI techniques was independently predictive of late HCC recurrence after liver resection. A risk score based on spleen volume, APRI score, and tumor number was developed for accurate prediction of late recurrence. KEY POINTS: • Preoperative spleen volume measured on MRI was independently associated with late recurrence after curative-intent resection in patients with HCC and cirrhosis. • Qualitative MRI features reflecting tumor aggressiveness were not associated with late recurrence. • A risk score based on spleen volume was developed for accurate prediction of late recurrence and risk stratification.

Preoperative MRI-based multiparametric model for survival prediction in hepatocellular carcinoma patients with portal vein tumor thrombus following hepatectomy.

PURPOSE: To develop a predictive model integrating clinical and MRI features for postoperative survival in patients with hepatocellular carcinoma (HCC) and portal vein tumor thrombus (PVTT). METHOD: Between January 2008 and May 2021, consecutive HCC patients with PVTT who underwent preoperative contrast-enhanced MRI and surgical resection at a tertiary hospital were retrospectively enrolled. The MR images were independently reviewed by two blinded radiologists. Univariate and multivariate Cox regression analyses were performed to construct a prognostic score for overall survival (OS). RESULTS: Ninety-four patients were included (mean age, 50.1 years; 84 men). During a median follow-up period of 15.3 months, 72 (76.6%) patients died (median OS, 15.4 months; median disease-free survival [DFS], 4.6 months). The sum size of the two largest tumors (hazard ratio [HR], 3.050; p < 0.001) and tumor growth subtype (HR, 1.928; p = 0.006) on MRI, serum albumin (HR, 0.948; p = 0.02), and age (HR, 0.978; p = 0.04) were associated with OS and incorporated in the prognostic score. Accordingly, patients were stratified into a high-risk or low-risk group, and the OS in the high-risk group was shorter than that in the low-risk group for the entire cohort (11.7 vs. 25.0 months, p < 0.001) and for patients with Cheng's type I (12.1 vs. 25.9 months, p = 0.002) and type II PVTT (11.7 vs. 25.0 months, p = 0.004). The DFS in the high-risk group was shorter than that in the low-risk group for the entire cohort (4.5 vs. 6.1 months, p = 0.001). CONCLUSIONS: Based on the sum size of the two largest tumors, tumor growth subtype, albumin, and age, the prognostic score allowed accurate preoperative risk stratification in HCC patients with PVTT, independent of Cheng's PVTT classification.

Diagnostic study on clinical feasibility of an AI-based diagnostic system as a second reader on mobile CT images: a preliminary result.

Background: Artificial intelligence (AI) has breathed new life into the lung nodules detection and diagnosis. However, whether the output information from AI will translate into benefits for clinical workflow or patient outcomes in a real-world setting remains unknown. This study was to demonstrate the feasibility of an AI-based diagnostic system deployed as a second reader in imaging interpretation for patients screened for pulmonary abnormalities in a clinical setting. Methods: The study included patients from a lung cancer screening program conducted in Sichuan Province, China using a mobile computed tomography (CT) scanner which traveled to medium-size cities between July 10th, 2020 and September 10th, 2020. Cases that were suspected to have malignant nodules by junior radiologists, senior radiologists or AI were labeled a high risk (HR) tag as HR-junior, HR-senior and HR-AI, respectively, and included into final analysis. The diagnosis efficacy of the AI was evaluated by calculating negative predictive value and positive predictive value when referring to the senior readers' final results as the gold standard. Besides, characteristics of the lesions were compared among cases with different HR labels. Results: In total, 251/3,872 patients (6.48%, male/female: 91/160, median age, 66 years) with HR lung nodules were included. The AI algorithm achieved a negative predictive value of 88.2% [95% confidence interval (CI): 62.2-98.0%] and a positive predictive value of 55.6% (95% CI: 49.0-62.0%). The diagnostic duration was significantly reduced when AI was used as a second reader (223±145.6 vs. 270±143.17 s, P<0.001). The information yielded by AI affected the radiologist's decision-making in 35/145 cases. Lesions of HR cases had a higher volume [309.9 (214.9-732.5) vs. 141.3 (79.3-380.8) mm3, P<0.001], lower average CT number [-511.0 (-576.5 to -100.5) vs. -191.5 (-487.3 to 22.5), P=0.010], and pure ground glass opacity rather than solid. Conclusions: The AI algorithm had high negative predictive value but low positive predictive value in diagnosing HR lung lesions in a clinical setting. Deploying AI as a second reader could help avoid missed diagnoses, reduce diagnostic duration, and strengthen diagnostic confidence for radiologists.

Comparison of a preoperative MR-based recurrence risk score versus the postoperative score and four clinical staging systems in hepatocellular carcinoma: a retrospective cohort study.

OBJECTIVES: To establish a risk score integrating preoperative gadoxetic acid-enhanced magnetic resonance imaging (EOB-MRI) and clinical parameters to predict recurrence after hepatectomy for patients with hepatocellular carcinoma (HCC) and to compare its performance with that of a postoperative score and four clinical staging systems. METHODS: Consecutive patients with surgically confirmed HCC who underwent preoperative EOB-MRI between July 2015 and November 2020 were retrospectively included. Two recurrence risk scores, one incorporating only preoperative variables and the other incorporating all preoperative and postoperative variables, were constructed via Cox regression models. RESULTS: A total of 214 patients (derivation set, n = 150; test set, n = 64) were included. Six preoperative variables, namely tumor number, infiltrative appearance, corona enhancement, alpha-fetoprotein (AFP) level, aspartate aminotransferase (AST) level, and sex, were independently associated with recurrence. After adding postoperative features, microvascular invasion and tumor differentiation were additional significant variables in lieu of corona enhancement and AFP level. Using the above variables, the preoperative score achieved a C-index of 0.741 on the test set, which was comparable with that of the postoperative score (0.729; p = 0.235). The preoperative score yielded a larger time-dependent area under the receiver operating characteristic curve at 1 year (0.844) than three existing systems (0.734-0.742; p < 0.05 for all). Furthermore, the preoperative score stratified patients into two prognostically distinct risk strata with low and high risks of recurrence (p < 0.001). CONCLUSION: The preoperative score integrating EOB-MRI features, AFP and AST levels, and sex improves recurrence risk estimation in HCC. KEY POINTS: • The preoperative risk score incorporating three EOB-MRI findings, AFP and AST levels, and sex achieved comparable performance with that of the postoperative score for predicting recurrence after hepatectomy in patients with HCC. • Two risk strata with low and high risks of recurrence were obtained based on the preoperative score. • The preoperative score may help tailor pretreatment decision-making and facilitate candidate selection for adjuvant clinical trials.

Magnetic resonance elastography biomarkers for detection of histologic alterations in nonalcoholic fatty liver disease in the absence of fibrosis.

OBJECTIVES: To investigate associations between histology and hepatic mechanical properties measured using multiparametric magnetic resonance elastography (MRE) in adults with known or suspected nonalcoholic fatty liver disease (NAFLD) without histologic fibrosis. METHODS: This was a retrospective analysis of 88 adults who underwent 3T MR exams including hepatic MRE and MR imaging to estimate proton density fat fraction (MRI-PDFF) within 180 days of liver biopsy. Associations between MRE mechanical properties (mean shear stiffness (|G*|) by 2D and 3D MRE, and storage modulus (G'), loss modulus (G″), wave attenuation (α), and damping ratio (ζ) by 3D MRE) and histologic, demographic and anthropometric data were assessed. RESULTS: In univariate analyses, patients with lobular inflammation grade ≥ 2 had higher 2D |G*| and 3D G″ than those with grade ≤ 1 (p = 0.04). |G*| (both 2D and 3D), G', and G″ increased with age (rho = 0.25 to 0.31; p ≤ 0.03). In multivariable regression analyses, the association between inflammation grade ≥ 2 remained significant for 2D |G*| (p = 0.01) but not for 3D G″ (p = 0.06); age, sex, or BMI did not affect the MRE-inflammation relationship (p > 0.20). CONCLUSIONS: 2D |G*| and 3D G″ were weakly associated with moderate or severe lobular inflammation in patients with known or suspected NAFLD without fibrosis. With further validation and refinement, these properties might become useful biomarkers of inflammation. Age adjustment may help MRE interpretation, at least in patients with early-stage disease. KEY POINTS: • Moderate to severe lobular inflammation was associated with hepatic elevated shear stiffness and elevated loss modulus (p =0.04) in patients with known or suspected NAFLD without liver fibrosis; this suggests that with further technical refinement these MRE-assessed mechanical properties may permit detection of inflammation before the onset of fibrosis in NAFLD. • Increasing age is associated with higher hepatic shear stiffness, and storage and loss moduli (rho = 0.25 to 0.31; p ≤ 0.03); this suggests that age adjustment may help interpret MRE results, at least in patients with early-stage NAFLD.

Diagnostic accuracy of hepatic proton density fat fraction measured by magnetic resonance imaging for the evaluation of liver steatosis with histology as reference standard: a meta-analysis.

OBJECTIVES: The aim of this meta-analysis was to evaluate the diagnostic accuracy of hepatic magnetic resonance imaging-proton density fat fraction (MRI-PDFF) for the assessment of liver steatosis (LS) with histology as reference standard. METHODS: A systematic literature search was performed to identify pertinent studies. Quality analyses were conducted by Quality Assessment of Diagnostic Accuracy Studies-2. Diagnostic data were extracted and inconsistency index was calculated for LS≥G1, LS≥G2, and LS=G3, respectively. The area under summary receiver operating characteristic curve (AUC) served as the indicator of diagnostic accuracy. The pooled sensitivity and specificity were calculated if threshold effect was absent. RESULTS: Thirteen studies containing 1100 subjects were included. There was significant threshold effect for LS≥G1. The AUCs for LS≥G1, LS≥G2, and LS=G3 were 0.98 (95% confidence interval (CI) 0.76, 1.00), 0.91 (95% CI 0.89, 0.94), and 0.92 (95% CI 0.89, 0.94), respectively. The pooled sensitivities for LS≥G2 and LS=G3 were 0.83 (95% CI 0.75, 0.88) and 0.79 (95% CI 0.63, 0.90), respectively; the pooled specificities for LS≥G2 and LS=G3 were 0.89 (95% CI 0.84, 0.92) and 0.89 (95% CI 0.84, 0.92), respectively. CONCLUSIONS: MRI-PDFF has high diagnostic accuracy at detecting and grading LS with histology as reference standard, suggesting that MRI-PDFF is able to provide an accurate quantification of LS in clinical trials and patient care. KEY POINT: • MRI-PDFF is able to provide an accurate quantification of LS in clinical trials and patient care.

Meta-analysis of dual-energy computed tomography virtual non-calcium imaging to detect bone marrow edema.

OBJECTIVE: This meta-analysis aimed to evaluate the accuracy of dual-energy CT (DECT) virtual non-calcium (VNC) imaging for the detection of bone marrow edema (BME). METHODS: A systematic literature search up to March 2017 was performed to find relevant original studies. Two reviewers independently selected studies, assessed literature quality, and extracted data. Pooled sensitivity, specificity, area under receiver operating characteristic (AUROC) curve, and other measures of DECT accuracy for detecting BME were calculated using random effects models. Risk of heterogeneity was assessed for the appropriateness of meta-analysis. RESULTS: Fourteen studies involving 2205 regions of vertebrae, hips, knees, and ankles were included. To evaluate the accuracy of BME detection using DECT, calculations were performed to obtain a pooled sensitivity of 0.812 (95% confidence interval [CI], 0.780-0.841) and specificity of 0.951 (95% CI, 0.940-0.960). The AUROC value was 0.9635. The major potential cause of heterogeneity was bone position. No significant publication bias was present. CONCLUSION: DECT VNC imaging gives very good diagnostic performance for BME detection and will likely be an important and common modality for acute assessment in the future.