Faculty Bibliography

BACKGROUND:To assess the performance of imaging features, including radiomics texture features, in predicting histopathologic tumor grade, AJCC stage, and outcomes [time to recurrence (TTR) and overall survival (OS)] in patients with intrahepatic cholangiocarcinoma (ICC). METHODS:Seventy-three patients (26 M/47F, mean age 63y) with pre-operative imaging (CT, n = 37; MRI, n = 21; CT and MRI, n = 15] within 6 months of resection were included in this retrospective study. Qualitative imaging traits were assessed by 2 observers. A 3rd observer measured tumor apparent diffusion coefficient (ADC), enhancement ratios (ERs), and Haralick texture features. Blood biomarkers and imaging features were compared with histopathology (tumor grade and AJCC stage) and outcomes (TTR and OS) using log-rank, generalized Wilcoxon, Cox proportional hazards regression, and Fisher exact tests. RESULTS:Median TTR and OS were 53.9 and 79.7 months. ICC recurred in 64.4% (47/73) of patients and 46.6% (34/73) of patients died. There was fair accuracy for some qualitative imaging features in the prediction of worse tumor grade (maximal AUC of 0.68 for biliary obstruction on MRI, p = 0.032, observer 1) and higher AJCC stage (maximal AUC of 0.73 for biliary obstruction on CT, p = 0.002, observer 2; and AUC of 0.73 for vascular involvement on MRI, p = 0.01, observer 2). Cox proportional hazards regression analysis showed that CA 19-9 [hazard ratio (HR) 2.44/95% confidence interval (CI) 1.31-4.57/p = 0.005)] and tumor size on imaging (HR 1.13/95% CI 1.04-1.22/p = 0.003) were significant predictors of TTR, while CA 19-9 (HR 4.08/95% CI 1.75-9.56, p = 0.001) and presence of metastatic lymph nodes at histopathology (HR 2.86/95% CI 1.35-6.07/p = 0.006) were significant predictors of OS. On multivariable analysis, satellite lesions on CT (HR 2.79/95%CI 1.01-7.15/p = 0.032, observer 2), vascular involvement on MRI (HR 0.10/95% CI 0.01-0.85/p = 0.032, observer 1), and texture feature MRI variance (HR 0.55/95% CI 0.31-0.97, p = 0.040) predicted TTR once adjusted for the independent predictors CA 19-9 and tumor size on imaging. Several qualitative and quantitative features demonstrated associations with TTR, OS, and AJCC stage at univariable analysis (range: HR 0.35-19; p < 0.001-0.045), however none were predictive of OS at multivariable analysis when adjusted for CA 19-9 and metastatic lymph nodes (p > 0.088). CONCLUSIONS:There was reasonable accuracy in predicting tumor grade and higher AJCC stage in ICC utilizing certain qualitative and quantitative imaging traits. Serum CA 19-9, tumor size, presence of metastatic lymph nodes, and qualitative imaging traits of satellite lesions and vascular involvement are predictors of patient outcomes, along with a promising predictive ability of certain quantitative texture features.

Background The methods for assessing knee osteoarthritis (OA) do not provide enough comprehensive information to make robust and accurate outcome predictions. Purpose To develop a deep learning (DL) prediction model for risk of OA progression by using knee radiographs in patients who underwent total knee replacement (TKR) and matched control patients who did not undergo TKR. Materials and Methods In this retrospective analysis that used data from the OA Initiative, a DL model on knee radiographs was developed to predict both the likelihood of a patient undergoing TKR within 9 years and Kellgren-Lawrence (KL) grade. Study participants included a case-control matched subcohort between 45 and 79 years. Patients were matched to control patients according to age, sex, ethnicity, and body mass index. The proposed model used a transfer learning approach based on the ResNet34 architecture with sevenfold nested cross-validation. Receiver operating characteristic curve analysis and conditional logistic regression assessed model performance for predicting probability and risk of TKR compared with clinical observations and two binary outcome prediction models on the basis of radiographic readings: KL grade and OA Research Society International (OARSI) grade. Results Evaluated were 728 participants including 324 patients (mean age, 64 years ± 8 [standard deviation]; 222 women) and 324 control patients (mean age, 64 years ± 8; 222 women). The prediction model based on DL achieved an area under the receiver operating characteristic curve (AUC) of 0.87 (95% confidence interval [CI]: 0.85, 0.90), outperforming a baseline prediction model by using KL grade with an AUC of 0.74 (95% CI: 0.71, 0.77; P < .001). The risk for TKR increased with probability that a person will undergo TKR from the DL model (odds ratio [OR], 7.7; 95% CI: 2.3, 25; P < .001), KL grade (OR, 1.92; 95% CI: 1.17, 3.13; P = .009), and OARSI grade (OR, 1.20; 95% CI: 0.41, 3.50; P = .73). Conclusion The proposed deep learning model better predicted risk of total knee replacement in osteoarthritis than did binary outcome models by using standard grading systems. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Richardson in this issue.

OBJECTIVES/OBJECTIVE:The primary objective was to compare the performance of 3 different abbreviated MRI (AMRI) sets extracted from a complete gadoxetate-enhanced MRI obtained for hepatocellular carcinoma (HCC) screening. Secondary objective was to perform a preliminary cost-effectiveness analysis, comparing each AMRI set to published ultrasound performance for HCC screening in the USA. METHODS:This retrospective study included 237 consecutive patients (M/F, 146/91; mean age, 58 years) with chronic liver disease who underwent a complete gadoxetate-enhanced MRI for HCC screening in 2017 in a single institution. Two radiologists independently reviewed 3 AMRI sets extracted from the complete exam: non-contrast (NC-AMRI: T2-weighted imaging (T2wi)+diffusion-weighted imaging (DWI)), dynamic-AMRI (Dyn-AMRI: T2wi+DWI+dynamic T1wi), and hepatobiliary phase AMRI (HBP-AMRI: T2wi+DWI+T1wi during the HBP). Each patient was classified as HCC-positive/HCC-negative based on the reference standard, which consisted in all available patient data. Diagnostic performance for HCC detection was compared between sets. Estimated set characteristics, including historical ultrasound data, were incorporated into a microsimulation model for cost-effectiveness analysis. RESULTS:The reference standard identified 13/237 patients with HCC (prevalence, 5.5%; mean size, 33.7 ± 30 mm). Pooled sensitivities were 61.5% for NC-AMRI (95% confidence intervals, 34.4-83%), 84.6% for Dyn-AMRI (60.8-95.1%), and 80.8% for HBP-AMRI (53.6-93.9%), without difference between sets (p range, 0.06-0.16). Pooled specificities were 95.5% (92.4-97.4%), 99.8% (98.4-100%), and 94.9% (91.6-96.9%), respectively, with a significant difference between Dyn-AMRI and the other sets (p < 0.01). All AMRI methods were effective compared with ultrasound, with life-year gain of 3-12 months against incremental costs of US$ < 12,000. CONCLUSIONS:NC-AMRI has limited sensitivity for HCC detection, while HBP-AMRI and Dyn-AMRI showed excellent sensitivity and specificity, the latter being slightly higher for Dyn-AMRI. Cost-effectiveness estimates showed that AMRI is effective compared with ultrasound. KEY POINTS/CONCLUSIONS:• Comparison of different abbreviated MRI (AMRI) sets reconstructed from a complete gadoxetate MRI demonstrated that non-contrast AMRI has low sensitivity (61.5%) compared with contrast-enhanced AMRI (80.8% for hepatobiliary phase AMRI and 84.6% for dynamic AMRI), with all sets having high specificity. • Non-contrast and hepatobiliary phase AMRI can be performed in less than 14 min (including set-up time), while dynamic AMRI can be performed in less than 17 min. • All AMRI sets were cost-effective for HCC screening in at-risk population in comparison with ultrasound.

PURPOSE/OBJECTIVE:To evaluate the correlation between liver stiffness as measured on MR elastography and T1 and T2 relaxation times from T1 and T2 mapping with clinical parameters of liver disease, including the MELD score, MELD-Na and ALBI grade, and endoscopically visible esophageal varices. MATERIALS AND METHODS/METHODS:223 patients with known or suspected liver disease underwent MRI of the liver with T1 mapping (Look-Locker sequence) and 2D SE-EPI MR elastography (MRE) sequences. 139 of these patients also underwent T2 mapping with radial T2 FS sequence. Two readers measured liver stiffness, T1 relaxation times and T2 relaxation times, and assessed qualitative features such as presence or absence of cirrhosis, ascites, spleen length, and varices on conventional MRI images. A third reader collected the clinical data (MELD score, MELD-Na Score, ALBI grade, and results of endoscopy in 78 patients). RESULTS:Significant moderate correlation was found between MELD score and all three imaging techniques for both readers (MRE, r = 0.35 and 0.28; T1 relaxometry, r = 0.30 and 0.29; T2 relaxometry, r = 0.45, and 0.37 for reader 1 and reader 2 respectively). Correlation with MELD-Na score was even higher (MRE, r = 0.49 and 0.40; T1, r = 0.45 and 0.41; T2, r = 0.47 and 0.35 for reader 1 and reader 2 respectively). Correlations between MRE and ALBI grade was significant and moderate for both readers: r = 0.39 and 0.37, higher than T1 relaxometry (r = 0.22 and 0.20) and T2 relaxometry (r = 0.17, and r = 0.24). Significant moderate correlations were found for both readers between MRE and the presence of varices on endoscopy (r = 0.28 and 0.30). MRE and T1 relaxometry were significant predictors of varices at endoscopy for both readers (MRE AUC 0.923 and 0.873; T1 relaxometry AUC = 0.711 and 0.675 for reader 1 and reader 2 respectively). Cirrhotic morphology (AUC = 0.654), spleen length (AUC = 0.610) and presence of varices in the upper abdomen on MRI (AUC of 0.693 and 0.595) were all significant predictors of endoscopic varices. Multivariable logistic regression model identified that spleen length and liver MRE were significant independent predictors of endoscopic varices for both readers. CONCLUSION/CONCLUSIONS:MR elastography, T1 and T2 relaxometry demonstrated moderate positive correlation with the MELD score and MELD-Na Score. Correlation between MRE and ALBI grade was superior to T1 and T2 relaxometry methods. MRE performed better than T1 and T2 relaxometry to predict the presence of varices at endoscopy. On multivariate analyses, spleen length and MRE were the only two significant independent predictors of endoscopic varices.

Purpose: To identify clinical, procedural, and logistical factors that influence time to inpatient interventional radiology (IR) procedures. Materials: All inpatient IR procedures performed at two tertiary care academic medical centers in January 2018 were retrospectively reviewed. Procedures were included if a complete consult note (with an associated time), and procedure start time were available. Time to procedure (TTP) was defined as the interval from consult note entry to procedure start time in hours. Clinical and procedure data which may influenced TTP were analyzed, including day of week, time of consult, procedure urgency and complexity, availability of imaging and laboratory values, requesting clinical service, patient vital signs, and procedural urgency. Consult time of day was divided into four time periods: early day (08:00-12:00), late day (12:00-16:00), evening (16:00-20:00), and overnight (20:00 - 08:00).
Result(s): A total of 127 inpatient procedures were performed on 116 patients (mean age, 59 years; 43% male). Procedures performed on Wednesdays and Fridays had the longest TTP (mean, 32 and 21 hours respectively, P = 0.010). Procedures performed during the weekend and on Mondays had the shortest TTP (mean, 2.9 and 10.8 hours, respectively, P = 0.010). The time of day the consult was completed correlated significantly with TTP (P = 0.038), with the shortest TTP for consults requested in the early day (mean, 11.4 hours) and overnight (mean, 11.5 hours) and the longest TTP for those requested in the afternoon (mean, 27.4 hours). Lack of appropriate imaging resulted in longer TTP (mean, 35 vs. 17 hours, P = 0.029). High urgency procedures had significantly shorter TTP (P = 0.038). There was no significant correlation between TTP and fasting status (P = 0.073), anticoagulation (P = 0.073), availability of appropriate labs (0.225), procedure category (P = 0.086), bed location (P = 0.094), and requesting service (P = 0.100).
Conclusion(s): Overnight, early day, and urgent procedures had the shortest TTP, whereas afternoon and later week consults had the longest TTP. Examining the underlying reasons for these trends may offer opportunities to reduce TTP for inpatient IR procedures.
Copyright

Beta amyloid (Aβ) accumulation is the earliest pathological marker of Alzheimer's disease (AD), but early AD pathology also affects white matter (WM) integrity. We performed a cross-sectional study including 44 subjects (23 healthy controls and 21 mild cognitive impairment or early AD patients) who underwent simultaneous PET-MR using 18F-Florbetapir, and were categorized into 3 groups based on Aβ burden: Aβ- [mean mSUVr ≤1.00], Aβi [1.00 < mSUVr <1.17], Aβ+ [mSUVr ≥1.17]. Intergroup comparisons of diffusion MRI metrics revealed significant differences across multiple WM tracts. Aβi group displayed more restricted diffusion (higher fractional anisotropy, radial kurtosis, axonal water fraction, and lower radial diffusivity) than both Aβ- and Aβ+ groups. This nonmonotonic trend was confirmed by significant continuous correlations between mSUVr and diffusion metrics going in opposite direction for 2 cohorts: pooled Aβ-/Aβi and pooled Aβi/Aβ+. The transient period of increased diffusion restriction may be due to inflammation that accompanies rising Aβ burden. In the later stages of Aβ accumulation, neurodegeneration is the predominant factor affecting diffusion.

PURPOSE/OBJECTIVE:To assess the diagnostic performance of magnetic resonance imaging (MRI) compared to blood tests and clinical scoring systems for the evaluation of histopathologic severity in patients with primary sclerosing cholangitis (PSC). MATERIALS/METHODS:Fifty-one patients (M/F 37/14, mean age 41 years) with PSC who underwent MRI and liver histopathology were included in this IRB-approved retrospective study. Two radiologists independently graded the severity of biliary abnormalities on magnetic resonance cholangiopancreatography (MRCP) using a standardized scoring system, parenchymal enhancement, and diffusion-weighted imaging (DWI) signal. Liver function tests, Mayo Risk score, APRI, FIB-4 Index, MELD, and Child-Pugh scores were recorded. Histopathology was assessed using a modified Nakanuma's scoring system. Correlation and diagnostic performance of MRI scores and blood tests for assessment of PSC histopathologic disease severity were evaluated. RESULTS:Findings of cirrhosis and portal hypertension were the only imaging features diagnostic of advanced PSC (stages 3 and 4) with AUC up to 0.90 (p < 0.001) for both observers. Parenchymal enhancement and overall qualitative biliary ductal abnormality identified advanced PSC stage with AUC up to 0.767 (p = 0.002) only for one observer. There was weak correlation between the overall qualitative biliary ductal abnormality on MRCP and histopathologic stage (r = 0.36, p = 0.01) for one observer. FIB-4 index, Child-Pugh, MELD, Mayo Risk, APRI, and alkaline phosphatase demonstrated good to excellent performance for advanced PSC stage (AUCs 0.672-0.915, p < 0.045). CONCLUSIONS:MRI findings of cirrhosis/portal hypertension, blood tests, and clinical scoring systems had high performance for advanced histopathologic PSC stage diagnosis, while the severity of biliary abnormalities on MRI did not.

Decreased brain lateralization is considered a trait marker of schizophrenia. Whereas reductions in both functional and macrostructural gray matter laterality in schizophrenia are well established, the investigation of gray matter microstructural lateralization has so far been limited to a small number of ex vivo studies, which limits the understanding of neurobiological substrates involved and development of adequate treatments. The aim of the current study was to assess in vivo gray matter microstructure lateralization patterns in schizophrenia by employing the diffusion kurtosis imaging (DKI)-derived mean kurtosis (MK) metric. MK was calculated for 18 right-handed males with chronic schizophrenia and 19 age-matched healthy control participants in 46 bilateral gray matter regions of interest (ROI). Microstructural laterality indexes (μLIs) were calculated for each subject and ROI, and group comparisons were conducted across regions. The relationship between μLI values and performance on the Wisconsin Card Sorting Test (WCST) was also evaluated. We found that compared with healthy controls, males with chronic schizophrenia had significantly decreased μLI across cortical and subcortical gray matter regions, which was correlated with poorer performance on the WCST. Our results suggest the ability of DKI-derived MK to capture gray matter microstructural lateralization pathology in vivo.

OBJECTIVE:The purpose of this study was to determine whether SWE can detect biomechanical changes in the supraspinatus muscle that occur with increasing supraspinatus tendon abnormality prior to morphologic gray-scale changes. MATERIALS AND METHODS/METHODS:An IRB approved, HIPAA compliant retrospective study of shoulder ultrasounds from 2013-2018 was performed. The cohort consisted of 88 patients (mean age 55 ± 15 years old) with 110 ultrasounds. Images were acquired in longitudinal orientation to the supraspinatus muscle with shear wave velocity (SWV) point quantification. The tendon and muscle were graded in order of increasing tendinosis/tear (1-4 scale) and increasing fatty infiltration (0-3 scale). Mixed model analysis of variance, analysis of covariance, and Spearman rank correlation were used for statistical analysis. RESULTS:There was no statistically significant age or sex dependence for supraspinatus muscle SWV (p = 0.314, 0.118, respectively). There was no significant correlation between muscle SWV and muscle or tendon grade (p = 0.317, 0.691, respectively). In patients with morphologically normal muscle on gray-scale ultrasound, there were significant differences in muscle SWV when comparing tendon grade 3 with grades 1, 2, and 4 (p = 0.018, 0.025, 0.014, respectively), even when adjusting for gender and age (p = 0.044, 0.028, 0.018, respectively). Pairwise comparison of tendon grades other than those mentioned did not achieve statistical significance (p > 0.05). CONCLUSION/CONCLUSIONS:SWE can detect biomechanical differences within the supraspinatus muscle that are not morphologically evident on gray-scale ultrasound. Specifically, supraspinatus tendon partial tears with moderate to severe tendinosis may correspond to biomechanically distinct muscle properties compared to both lower grades of tendon abnormality and full-thickness tears.

OBJECTIVES/OBJECTIVE:To evaluate the diagnostic performance of contrast-enhanced CT vs. MRI with extracellular contrast agents (EC-MRI) vs. MRI with gadoxetic acid (EOB-MRI) for HCC detection in patients with liver cirrhosis using liver explant as the reference. The additional value of hepatobiliary phase (HBP) post Gadoxetic acid was also assessed. METHODS:Two-hundred seventy-seven consecutive patients who underwent liver transplantation over a 9 year period and imaging within 90 days of were retrospectively included. Imaging consisted in CT (n = 100), EC-MRI (n = 77) and EOB-MRI (n = 100), the latter subdivided into dynamic EOB-MRI and full EOB-MRI (dynamic+HBP). Three radiologists retrospectively categorized lesions ≥ 1 cm using the LI-RADSv2017 algorithm. Dynamic EOB-MRI was re-evaluated with the addition of HBP. Results were correlated with explant pathology. RESULTS:Pathology demonstrated 265 HCCs (mean size 2.1 ± 1.4 cm) in 177 patients. Per-patient sensitivities were 86.3% for CT, 89.5% for EC-MRI, 92.8% for dynamic EOB-MRI and 95.2% for full EOB-MRI (pooled reader data), with a significant difference between CT and dynamic/full EOB-MRI (p = 0.032/0.002), and between EC-MRI and full EOB-MRI (p = 0.047). Per-lesion sensitivities for CT, EC-MRI, dynamic EOB-MRI and full EOB-MRI were 59.5%,78.5%,69.7% and 76.8%, respectively, with a significant difference between MRI groups and CT (p-range:0.001-0.04), and no difference between EC-MRI and dynamic EOB-MRI (p = 0.949). For HCCs 1-1.9 cm, sensitivities were 34.4%, 64.6%, 57.3% and 67.3%, respectively, with all MRI groups significantly superior to CT (p ≤ 0.01) and full EOB-MRI superior to dynamic EOB-MRI (p = 0.002). CONCLUSIONS:EOB-MRI outperforms CT and EC-MRI for per-patient HCC detection sensitivity, and is equivalent to EC-MRI for per-lesion sensitivity. MRI methods outperform CT for detection of HCCs 1-1.9 cm. KEY POINTS/CONCLUSIONS:• MRI is superior to CT for HCC detection in patients with liver cirrhosis. • EOB-MRI outperforms CT and MRI using extracellular contrast agents (EC-MRI) for per-patient HCC detection sensitivity, and is equivalent to EC-MRI for per-lesion sensitivity. • The addition of hepatobiliary phase images improves HCC detection when using gadoxetic acid.