Strategies for Implementing Machine Learning Algorithms in the Clinical Practice of Radiology.

Despite recent advancements in machine learning (ML) applications in health care, there have been few benefits and improvements to clinical medicine in the hospital setting. To facilitate clinical adaptation of methods in ML, this review proposes a standardized framework for the step-by-step implementation of artificial intelligence into the clinical practice of radiology that focuses on three key components: problem identification, stakeholder alignment, and pipeline integration. A review of the recent literature and empirical evidence in radiologic imaging applications justifies this approach and offers a discussion on structuring implementation efforts to help other hospital practices leverage ML to improve patient care. Clinical trial registration no. 04242667 © RSNA, 2024 Supplemental material is available for this article.

Evaluating outcomes and toxicities for a newly implemented MRI-based brachytherapy program for cervical cancer.

OBJECTIVE: We report an updated analysis of the outcomes and toxicities of MRI-based brachytherapy for locally advanced cervical cancer from a U.S. academic center. METHODS: A retrospective review was performed on patients treated with MRI-based brachytherapy for cervical cancer. EBRT was standardly 45 Gy in 25 fractions with weekly cisplatin. MRI was performed with the brachytherapy applicator in situ. Dose specification was most commonly 7 Gy for 4 fractions with optimization aim of D90 HR-CTV EQD2 of 85-95 Gyα/ß=10 Gy in 2 implants each delivering 2 fractions. RESULTS: Ninety-eight patients were included with median follow up of 24.5 months (IQR 11.9-39.8). Stage IIIA-IVB accounted for 31.6% of cases. Dosimetry results include median GTV D98 of 101.0 Gy (IQR 93.3-118.8) and HR-CTV D90 of 89 Gy (IQR 86.1-90.6). Median D2cc bladder, rectum, sigmoid, and bowel doses were 82.1 Gy (IQR 75.9-88.0), 65.9 Gy (IQR 59.6-71.2), 65.1 Gy (IQR 57.7-69.6), and 55 Gy (IQR 48.9-60.9). Chronic grade 3+ toxicities were seen in the bladder (8.2%), rectosigmoid (4.1%), and vagina (1.0%). Three-year LC, PFS, and OS were estimated to be 84%, 61.7%, and 76.1%, respectively. CONCLUSION: MRI-based brachytherapy demonstrates excellent local control and acceptable rates of high-grade morbidity. These results are possible in our population with relatively large volume primary tumors and extensive local disease.

A prostate biopsy risk calculator based on MRI: development and comparison of the Prospective Loyola University multiparametric MRI (PLUM) and Prostate Biopsy Collaborative Group (PBCG) risk calculators.

OBJECTIVES: To develop and validate a prostate cancer (PCa) risk calculator (RC) incorporating multiparametric magnetic resonance imaging (mpMRI) and to compare its performance with that of the Prostate Biopsy Collaborative Group (PBCG) RC. PATIENTS AND METHODS: Men without a PCa diagnosis receiving mpMRI before biopsy in the Prospective Loyola University mpMRI (PLUM) Prostate Biopsy Cohort (2015-2020) were included. Data from a separate institution were used for external validation. The primary outcome was diagnosis of no cancer, grade group (GG)1 PCa, and clinically significant (cs)PCa (≥GG2). Binary logistic regression was used to explore standard clinical and mpMRI variables (prostate volume, Prostate Imaging-Reporting Data System [PI-RADS] version 2.0 lesions) with the final PLUM RC, based on a multinomial logistic regression model. Receiver-operating characteristic curve, calibration curves, and decision-curve analysis were evaluated in the training and validation cohorts. RESULTS: A total of 1010 patients were included for development (N = 674 training [47.8% PCa, 30.9% csPCa], N = 336 internal validation) and 371 for external validation. The PLUM RC outperformed the PBCG RC in the training (area under the curve [AUC] 85.9% vs 66.0%; P < 0.001), internal validation (AUC 88.2% vs 67.8%; P < 0.001) and external validation (AUC 83.9% vs 69.4%; P < 0.001) cohorts for csPCa detection. The PBCG RC was prone to overprediction while the PLUM RC was well calibrated. At a threshold probability of 15%, the PLUM RC vs the PBCG RC could avoid 13.8 vs 2.7 biopsies per 100 patients without missing any csPCa. At a cost level of missing 7.5% of csPCa, the PLUM RC could have avoided 41.0% (566/1381) of biopsies compared to 19.1% (264/1381) for the PBCG RC. The PLUM RC compared favourably with the Stanford Prostate Cancer Calculator (SPCC; AUC 84.1% vs 81.1%; P = 0.002) and the MRI-European Randomized Study of Screening for Prostate Cancer (ERSPC) RC (AUC 84.5% vs 82.6%; P = 0.05). CONCLUSIONS: The mpMRI-based PLUM RC significantly outperformed the PBCG RC and compared favourably with other mpMRI-based RCs. A large proportion of biopsies could be avoided using the PLUM RC in shared decision making while maintaining optimal detection of csPCa.

Risk of prostate cancer for men with prior negative biopsies undergoing magnetic resonance imaging compared with biopsy-naive men: A prospective evaluation of the PLUM cohort.

BACKGROUND: Men with prior negative prostate biopsies have a lower risk of being diagnosed with prostate cancer in comparison with biopsy-naive men. However, the relative clinical utility of identified lesions on multiparametric magnetic resonance imaging (mpMRI) is uncertain between the 2 settings. METHODS: Patients from the Prospective Loyola University mpMRI (PLUM) Prostate Biopsy Cohort (January 2015 to June 2020) were examined. The detection of any prostate cancer and clinically significant prostate cancer (Gleason score ≥ 3 + 4) was stratified by Prostate Imaging-Reporting and Data System (PI-RADS) scores in the prior negative and biopsy-naive settings. Multivariable logistic regression models (PLUM models) assessed predictors, and decision curve analyses were used to estimate the clinical utility of PI-RADS cutoffs relative to the models. RESULTS: Nine hundred men (420 prior negative patients and 480 biopsy-naive patients) were included. Prior negative patients had lower risks of any prostate cancer (27.9% vs 54.4%) and clinically significant prostate cancer (20.0% vs 38.3%) in comparison with biopsy-naive patients, and this persisted when they were stratified by PI-RADS (eg, PI-RADS 3: 13.6% vs 27.4% [any prostate cancer] and 5.2% vs 15.4% [clinically significant prostate cancer]). The rate of detection of clinically significant prostate cancer was 5.3% among men with prior negative biopsy and PI-RADS ≤ 3. Family history and Asian ancestry were significant predictors among biopsy-naive patients. PLUM models demonstrated a greater net benefit and reduction in biopsies (45.8%) without missing clinically significant cancer in comparison with PI-RADS cutoffs (PI-RADS 4: 34.0%). CONCLUSIONS: Patients with prior negative biopsies had lower prostate cancer detection by PI-RADS score category in comparison with biopsy-naive men. Decision curve analyses suggested that many biopsies could be avoided by the use of the PLUM models or a PI-RADS 4 cutoff without any clinically significant cancer being missed. LAY SUMMARY: Men with a prior negative prostate biopsy had a lower risk of harboring prostate cancer in comparison with those who never had a biopsy. This was true even when patients in each group had similar multiparametric magnetic resonance imaging (mpMRI) findings in terms of Prostate Imaging-Reporting and Data System (PI-RADS)-graded lesions. Decision curve analyses showed that many biopsies could be avoided by the use of the Prospective Loyola University mpMRI prediction models or a PI-RADS 4 cutoff for patients with prior negative biopsies.

Systematic versus Targeted Magnetic Resonance Imaging/Ultrasound Fusion Prostate Biopsy among Men with Visible Lesions.

PURPOSE: Multiparametric magnetic resonance imaging (mpMRI)-ultrasound (US) fusion-guided biopsy may improve prostate cancer (PCa) detection and reduce grade misclassification. We compared PCa detection rates on systematic, magnetic resonance imaging-targeted, and combined biopsy with evaluation of important subgroups. MATERIALS AND METHODS: Men with clinical suspicion of harboring PCa from 2 institutions with visible Prostate Imaging-Reporting and Data System (PI-RADSTMv2) lesions receiving mpMRI-US fusion-guided prostate biopsy were included (2015-2020). Detection of PCa was categorized by grade group (GG). Clinically-significant PCa (csPCa) was defined as ≥GG2. Patients were stratified by biopsy setting and PI-RADS. RESULTS: Of 1,236 patients (647 biopsy-naïve) included, 626 (50.6%) harbored PCa and 412 (33.3%) had csPCa on combined biopsy. Detection of csPCa was 27.9% vs 23.3% (+4.6%) and GG1 PCa was 11.3% vs 17.8% (-6.5%) for targeted vs systematic cores. Benefit in csPCa detection was higher in the prior negative than biopsy-naïve setting (+7.8% [p <0.0001] vs +1.7% [p=0.3]) while reduction in GG1 PCa detection remained similar (-5.6% [p=0.0002] vs -7.3% [p=0.0001]). Targeted biopsy showed increased csPCa detection for PI-RADS 5, decrease in GG1 for PI-RADS 3, and both for PI-RADS 4 relative to systematic biopsy. Combined biopsy detected more csPCa (+10.0%) and slightly fewer GG1 PCa (-0.5%) compared to systematic alone. Upgrading to ≥GG2 by targeted biopsy occurred in 9.8% with no cancer and 23.6% with GG1 on systematic biopsy. CONCLUSIONS: Combined biopsy doubled the benefit of targeted biopsy alone in detection of csPCa without increasing GG1 PCa diagnoses relative to systematic biopsy. Utility of targeted biopsy was higher in the prior negative biopsy cohort, but advantages of combined biopsy were maintained regardless of biopsy history.

Validation of Second-Generation Motion-Correction Software for Computed Tomography Coronary Angiography With Novel Quantitative Approach.

INTRODUCTION: Computed tomography of the coronary arteries (CTCA) is an important diagnostic tool. However, motion degradation is sometimes a challenge to interpretation and quantification, particularly with elevated heart rates. Here, a novel quantitative method is presented as part of an evaluation of one particular motion correction algorithm. METHODS: Computed tomography of the coronary arteries scans from 49 patients, with heart rates of >70 bpm, were identified with motion artifacts in multiple coronary segments. At these foci (196), an objective measure of motion degradation, defined here by cross-section eccentricity, was determined before and after image processing with second-generation GE SnapShot Freeze software (SSF-2.0). In addition, a subjective scoring was applied by an expert cardiothoracic radiologist both before and after processing. RESULTS: An overall decrease in vessel eccentricity strongly correlated (P < 0.001) with processing of the images by motion-correction software. A concurrent overall increase in subjective vessel clarity correlated (P < 0.001) with application of the software as well. CONCLUSIONS: A novel quantitative method (and subjective analysis) for evaluation of CTCA motion has been described and applied to validation of SSF-2.0 motion-correction software. Both the technique and software demonstrate promise for robust clinical utility in CTCA evaluation.

Preserving the organ donor pool and suprahepatic vena cava: Case series of transverse hepatectomy for polycystic liver disease.

Management of symptomatic polycystic liver disease (PLD) has remained primarily unchanged since the early 20th century when multiple case reports described management of non-parasitic liver cysts. In 1968, Lin et al. described the fenestration procedure, "aspiration of the cysts, incision, partial excision with or without external drainage, or marsupilization and anastomosis to the gastrointestinal tract". Further surgical options have included cyst sclerotherapy, laparoscopic cyst aspiration, partial hepatectomy, and orthotopic liver transplant (OLT). Recently there has been discussion of medical management with somatostatin analogs to reduce hepatomegaly in PLD with varying success. There is no current consensus on treatment or standard of care for symptomatic PLD, it is largely up to surgeon preference and ability; however, there has been a movement toward early OLT with Model for End-Stage Liver Disease (MELD) score exception points. This case series reviews two female patients with normal renal and hepatic function with symptomatic PLD treated with transverse hepatectomy. We propose that patients suffering from symptomatic PLD, with retained renal and hepatic function, can be treated with transverse hepatectomy conserving limited donor livers for decompensated patients; moreover, transverse hepatectomy does not disrupt the major suprahepatic vena cava preserving potential surgical access for future OLT.

Variability in prostate cancer detection among radiologists and urologists using MRI fusion biopsy.

Objectives: The aim of this study is to evaluate the impact of radiologist and urologist variability on detection of prostate cancer (PCa) and clinically significant prostate cancer (csPCa) with magnetic resonance imaging (MRI)-transrectal ultrasound (TRUS) fusion prostate biopsies. Patients and methods: The Prospective Loyola University MRI (PLUM) Prostate Biopsy Cohort (January 2015 to December 2020) was used to identify men receiving their first MRI and MRI/TRUS fusion biopsy for suspected PCa. Clinical, MRI and biopsy data were stratified by radiologist and urologist to evaluate variation in Prostate Imaging-Reporting and Data System (PI-RADS) grading, lesion number and cancer detection. Multivariable logistic regression (MVR) models and area under the curve (AUC) comparisons assessed the relative impact of individual radiologists and urologists. Results: A total of 865 patients (469 biopsy-naïve) were included across 5 urologists and 10 radiologists. Radiologists varied with grading 15.4% to 44.8% of patients with MRI lesions as PI-RADS 3. PCa detection varied significantly by radiologist, from 34.5% to 66.7% (p = 0.003) for PCa and 17.2% to 50% (p = 0.001) for csPCa. Urologists' PCa diagnosis rates varied between 29.2% and 55.8% (p = 0.013) and between 24.6% and 39.8% (p = 0.36) for csPCa. After adjustment for case-mix on MVR, a fourfold to fivefold difference in PCa detection was observed between the highest-performing and lowest-performing radiologist (OR 0.22, 95%CI 0.10-0.47, p < 0.001). MVR demonstrated improved AUC for any PCa and csPCa detection when controlling for radiologist variation (p = 0.017 and p = 0.038), but controlling for urologist was not significant (p = 0.22 and p = 0.086). Any PCa detection (OR 1.64, 95%CI 1.06-2.55, p = 0.03) and csPCa detection (OR 1.57, 95%CI 1.00-2.48, p = 0.05) improved over time (2018-2020 vs. 2015-2017). Conclusions: Variability among radiologists in PI-RADS grading is a key area for quality improvement significantly impacting the detection of PCa and csPCa. Variability for performance of MRI-TRUS fusion prostate biopsies exists by urologist but with less impact on overall detection of csPCa.

MRI versus non-MRI diagnostic pathways before radical prostatectomy: Impact on nerve-sparing, positive surgical margins, and biochemical recurrence.

PURPOSE: Magnetic resonance imaging (MRI) prior to biopsy has improved detection of clinically significant prostate cancer (CaP), but its impact on surgical outcomes is less well established. We compared MRI vs. non-MRI diagnostic pathways among patients receiving radical prostatectomy (RP) for impact on surgical outcomes. MATERIALS AND METHODS: Men diagnosed with CaP and receiving RP at Loyola University Medical Center (2014-2021) were categorized into MRI or non-MRI diagnostic pathways based on receipt of MRI before prostate biopsy. Primary outcomes of interest included positive surgical margin (PSM) rates, the performance of bilateral nerve-sparing, and biochemical recurrence (BCR). Multivariable logistic regression models, Kaplan-Meier curves, and Cox proportional hazards regression were employed. RESULTS: Of 609 patients, 281 (46.1%) were in the MRI and 328 (53.9%) in the non-MRI groups. MRI patients had similar PSA, biopsy grade group (GG) distribution, RP GG, pT stage, and RP CaP volume compared to non-MRI patients. PSM rates were not statistically different for the MRI vs. non-MRI groups (22.8% vs. 26.8%, P = 0.25). Bilateral nerve-sparing rates were higher for the MRI vs. non-MRI groups (OR 1.95 (95%CI 1.32-2.88), P = 0.001). The MRI group demonstrated improved BCR (HR 0.64 (95%CI 0.41-0.99), P = 0.04) after adjustment for age, PSA, RP GG, pT, pN, and PSM status. On meta-analysis, a 5.2% PSM reduction was observed but high heterogeneity for use of nerve-sparing. CONCLUSIONS: An MRI-based diagnostic approach selected patients for RP with a small reduction in PSM rates, greater utilization of bilateral nerve-sparing, and improved cancer control by BCR compared to a non-MRI approach even after adjustment for known prognostic factors.

MRI vs Transrectal Ultrasound to Estimate Prostate Volume and PSAD: Impact on Prostate Cancer Detection.

OBJECTIVES: To compare multiparametric magnetic resonance imaging (mpMRI) and transrectal ultrasound (TRUS) to estimate prostate volume and prostate specific antigen density (PSAD) as well as subsequent impact on prostate cancer (PCa) detection. METHODS: Patients referred for mpMRI prior to mpMRI-TRUS fusion-guided prostate biopsy between 2015 and 2020 were identified. Volume and calculated PSAD by mpMRI and TRUS were compared. Associations with presence of any PCa and clinically significant PCa (csPCa; Gleason ≥3 + 4) were evaluated using linear regression (interaction by volume quartile), logistic regression, and receiver operating characteristics. RESULTS: Among 640 men, TRUS underestimated prostate volume relative to mpMRI (median 49.2cc vs. 54.1cc) with 8% lower volume per cc up to 77.5cc (First-third quartile) and 39% lower volume per additional cc above 77.5cc (fourth quartile). For men undergoing radical prostatectomy, mpMRI had a higher correlation coefficient relative to TRUS (0.913 vs 0.878) when compared to surgical pathology. mpMRI PSAD had slightly higher odds vs TRUS PSAD for detecting any PCa (OR 2.94 and OR 2.78, both P <.001) or csPCa (OR 4.20 and OR 4.02, both P <.001). AUC improvements were of borderline significance for mpMRI vs. TRUS PSAD for any PCa (0.689 vs 0.675, P = .05) and not significant for csPCa (0.732 vs 0.722, P = .20). PSAD was not associated with PCa detection for prostates ≥77.5cc. CONCLUSION: TRUS underestimates prostate volume relative to mpMRI. PSAD based on mpMRI may be better associated with detection of PCa compared to TRUS, but utility of PSAD may be limited for larger prostates.

Safety profile of Gadoterate meglumine on the renal function of patients with severe kidney disease.

OBJECTIVE: Intravenous administration of gadolinium-based contrast agents (GBCA) in patients with impaired renal function has been of concern to primary care physicians due to the potential worsening of renal dysfunction and nephrogenic systemic fibrosis (NSF). Our objective was to compare the potential change in estimated glomerular filtration rate (eGFR) in patients with known severe renal dysfunction (eGFR <30 ml/min), following Gadoterate meglumine (GM) administration with patients who do not receive contrast. METHODS: An IRB-approved retrospective analysis of all patients who underwent MRI examination at our institution, for any indication, between January 2016 and September 2020. INCLUSION CRITERIA: pre-MRI eGFR <30 ml/min within 24 h of MRI, follow-up eGFR between 48 and 96 h post-MRI, and absence of peritoneal or hemodialysis. The individuals who received GM (492 scans) were identified as cases, and those who did not receive contrast (1101 scans) were identified as controls for our study. Delta-eGFR response was calculated and covariate-adjusted, and propensity score analysis was performed. RESULTS: No significant eGFR decrease was observed in patients who received GM compared to those who did not receive GM in our study. Also, no relationship between comorbidity, severity and contrast selection was observed. CONCLUSION: The use of Gadolinium contrast in MRI is often of critical importance for determining accurate anatomic relationships, differentiation of benign from malignant lesions, or determination of resolving vs. worsening disease. Though the risk of contrast administration can never be entirely ignored, especially in patients with low eGFR, our study indicates that safe administration of GM can be performed even in patients with severe kidney disease.

Impact of late gadolinium enhancement extent, location, and pattern on ventricular tachycardia and major adverse cardiac events in patients with ischemic vs. non-ischemic cardiomyopathy.

Background: Left ventricular late gadolinium enhancement (LGE) by cardiac magnetic resonance (CMR) has been associated with increased risk for life-threatening ventricular tachyarrhythmias. The differences in association between LGE characteristics and prognosis in patients with ischemic (ICM) vs. non-ischemic (NICM) cardiomyopathy is incompletely understood. Methods: A total of 168 consecutive patients who underwent CMR imaging with either ICM or NICM were included in our study. LGE extent, location and pattern were examined for association to the primary endpoint of ventricular tachycardia (VT) and secondary endpoint of major adverse cardiac events (MACE). Results: Of 68 (41%) patients with ICM and 97 (59%) patients with NICM, median LGE mass was 15% (IQR 9-28) for the ICM group and 10% (IQR 6-15) for the NICM group. On multivariate analysis for both groups, LGE characteristics were prognostic while LVEF was not. In patients with ICM, septal and apical segment LGE, and involvement of multiple walls predicted both endpoints on multivariate analysis. LGE extent (≥median) and inferior wall LGE independently predicted the primary endpoint. In patients with NICM, anterior, inferior and apical segment LGE, and involvement of multiple walls predicted both endpoints on multivariate analysis. LGE extent (≥median, number of LGE segments, LGE stratified per 5% increase) and midwall LGE were independent predictors of the primary endpoint. Conclusions: Although LGE was an independent predictor of prognosis in both groups, LGE extent, location, and pattern characteristics were more powerful correlates to worse outcomes in patients with NICM than ICM.

Image and structured data analysis for prognostication of health outcomes in patients presenting to the ED during the COVID-19 pandemic.

BACKGROUND: Patients admitted to the emergency department (ED) with COVID-19 symptoms are routinely required to have chest radiographs and computed tomography (CT) scans. COVID-19 infection has been directly related to the development of acute respiratory distress syndrome (ARDS) and severe infections could lead to admission to intensive care and increased risk of death. The use of clinical data in machine learning models available at time of admission to ED can be used to assess possible risk of ARDS, the need for intensive care (admission to the Intensive Care Unit; ICU) as well as risk of mortality. In addition, chest radiographs can be inputted into a deep learning model to further assess these risks. PURPOSE: This research aimed to develop machine and deep learning models using both structured clinical data and image data from the electronic health record (EHR) to predict adverse outcomes following ED admission. MATERIALS AND METHODS: Light Gradient Boosting Machine (LightGBM) was used as the main machine learning algorithm using all clinical data including 42 variables. Compact models were also developed using the 15 most important variables to increase applicability of the models in clinical settings. To predict risk (or early stratified risk) of the aforementioned health outcome events, transfer learning from the CheXNet model was also implemented on the available data. This research utilized clinical data and chest radiographs of 3,571 patients, 18 years and older, admitted to the emergency department between 9th March 2020 and 29th October 2020 at Loyola University Medical Center. MAIN FINDINGS: The research results show that we can detect COVID-19 infection (AUC = 0.790 (0.746-0.835)), predict the risk of developing ARDS (AUC = 0.781 (0.690-0.872), risk stratification of the need for ICU admission (AUC = 0.675 (0.620-0.713)) and mortality (AUC = 0.759 (0.678-0.840)) at moderate accuracy from both chest X-ray images and clinical data. PRINCIPAL CONCLUSIONS: The results can help in clinical decision making, especially when addressing ARDS and mortality, during the assessment of patients admitted to the ED with or without COVID-19 symptoms.

Magnetization transfer ratio mapping of intervertebral disc degeneration.

The magnetization transfer ratio of the lumbar discs was spatially quantified from age-matched subjects and the nucleus pulposus magnetization transfer ratio was correlated with T2-weighted Pfirrmann grades. A moderate and significant linear correlation between magnetization transfer ratio and Pfirrmann grades was observed, suggesting that nucleus pulposus collagen relative density increases with degeneration. High-resolution axial magnetization transfer ratio maps revealed elevated magnetization transfer ratio in the nucleus pulposa of injured and heavily degenerated discs. In the injured disc, significant elevation in nucleus pulposa magnetization transfer ratio was not accompanied by significant decrease in disc height. This observation may suggest a possible increase in absolute collagen content, in addition to increased collagen relative density. In summary, magnetization transfer MRI of the disc may serve as a noninvasive diagnostic tool for disc degeneration, in addition to other MRI techniques specific to proteoglycan content.

Comparison of computed tomography hepatic steatosis criteria for identification of abnormal liver function and clinical risk factors, in incidentally noted fatty liver.

OBJECTIVES: Hounsfield Units (HU) to compare the various computed tomography (CT) criteria for diagnosing hepatic steatosis with laboratory liver function parameters, and clinical risk factors retrospectively, when hepatic steatosis was incidentally detected. METHODS: Institutional review board-approved, Health Insurance Portability and Accountability Act-compliant, retrospective study in 200 randomly selected patients who had either nonenhanced CT (NECT) or contrast-enhanced CT (CECT) studies with reported hepatic steatosis. The participants were matched to age, gender, and ethnicity with 200 patients without hepatic steatosis. For NECT, four different criteria have been proposed in the literature to diagnose fatty liver: (1) liver HU less than 48 HU; (2) ratio of liver to spleen HU less than 0.8; (3) HU difference between liver and spleen less than -10; and (4) hepatic vessel HU ≥ liver HU. For CECT, difference between liver and spleen HU, in portal venous phase, ≤ -20 to -25 HU. Serum glucose, aspartate aminotransferase (AST), amino alanine transferase (ALT), total bilirubin were documented. Clinical history and clinical risk factors were documented from the electronic health records. Matched analyses and Wilcoxon signed rank sum test analysis were performed for matched variables. RESULTS: Fatty liver by NECT criteria 1 and 3 has statistically significant correlation with elevated glucose levels (P = 0.02). Similarly, fatty liver by 1, 3, and 4 NECT criteria showed statistically significant associations with higher levels of ALT and AST. There were statistically significant higher prevalence of diabetes mellitus (P = 0.003) and alcohol consumption (P ≤ 0.0001) in cases when compared with the controls. There was marginal significance in CT Dose Index between cases and controls (95% confidence interval: 0.98, 1.00; odds ratio 0.99), reflecting that cases had slightly higher BMI compared to their matched controls, thereby requiring slightly higher mA/mAs for imaging. CONCLUSION: Particular NECT criteria for fatty liver are best at identification of abnormal liver function and certain comorbidities, in the setting of incidental fatty liver detection, This creates the potential for benefits of early detection in clinical management.

Role of Multiparametric Magnetic Resonance Imaging Prostate Specific Antigen Density and PI-RADS™ Score in Predicting Up Staging in Men on Active Surveillance.

INTRODUCTION: Using a combination of magnetic resonance imaging of the prostate and prostate specific antigen density, we determined which men on active surveillance are at risk for up staging and which men could avoid repeat biopsy while remaining on surveillance. METHODS: We reviewed the records of 110 men on active surveillance with Gleason 6 disease who underwent magnetic resonance imaging followed by UroNav fusion biopsy (Invivo, Gainesville, Florida). Using univariable and multivariable logistic regression analyses we examined the effect of age, race, prostate specific antigen, prostate specific antigen density, prostate volume, PI-RADS (Prostate Imaging Reporting and Data System) score, number and size of target lesions, and time on surveillance to determine the likelihood of up staging to Gleason 7 or greater disease. RESULTS: A total of 33 cases (30%) were up staged. On multivariable analysis prostate specific antigen density and PI-RADS score were significant predictors of up staging with adjusted odds ratios of 3.97 for prostate specific antigen density of 0.16 or more (CI 1.31-12.00, p <0.05), 13.8 for a PI-RADS 4 lesion (CI 2.3-81.3, p <0.01) and 25 for a PI-RADS 5 lesion (CI 3.8-163.5, p <0.01). When cross-tabulating these factors, men with a PI-RADS score of 3 or greater with a prostate specific antigen density of 0.16 ng/ml/cc or more had a 61.2% chance of up staging. Conversely, in men with PI-RADS score 3 or less and prostate specific antigen density less than 0.15, no up staging was seen. CONCLUSIONS: A combination of PI-RADS score and prostate specific antigen density predicts patients at risk for up staging at surveillance biopsy. Conversely, this combination may help determine which men may safely forgo biopsy.

Early outcomes and impact of a hybrid IC/IS applicator for a new MRI-based cervical brachytherapy program.

PURPOSE: The purpose of this study was to report early outcomes and assess the learning curve in a new MRI-based cervical brachytherapy program. METHODS: We accrued 33 patients prospectively, and only patients with ≥3 months' followup (n = 27) were assessed for disease control and toxicity. Eras were defined as first half and second half for the intracavitary (IC)-only era (n = 13 each), and the intracavitary/interstitial (IC/IS) era was separated by difference in applicator availability (n = 7). Dose to 90% of the high-risk clinical target volume (D90 HR-CTV) and minimum dose to the maximally irradiated 2 cubic centimeters (D2cc) to organs at risk were used to assess dosimetry. Statistics were performed with t tests and Kaplan-Meier method. RESULTS: Median followup was 14.7 months. Median treatment duration was 50.5 vs. 57 days for patients treated with external beam radiation therapy at our institution vs. an outside institution (p = 0.03). One-year local control, noncervical pelvic control, distant metastasis-free rate, and overall survival were 84.0%, 96.0%, 78.5%, and 91.3%, respectively. When comparing the first half and second half eras of IC only, there were no differences in median D90 HR-CTV or D2cc of the bladder, rectum, or sigmoid. Comparing the entire IC era to the IC/IS era, median D90 HR-CTV trended higher from 88.0 Gy to 92.9 Gy (p = 0.11). D2cc rectum decreased from 69.3 Gy to 62.6 Gy (p = 0.01), and D2cc bladder trended lower from 87.5 Gy to 83.6 Gy (p = 0.09). CONCLUSIONS: There was no significant difference between the first half and second half eras with IC-only MRI-based brachytherapy. Incorporation of an IC/IS applicator generated the greatest dosimetric improvement. Early results of the MRI-based brachytherapy program are favorable.

Comparison of correlations between lipid profile and different computed tomography fatty liver criteria in the setting of incidentally noted fatty liver on computed tomography examinations.

PURPOSE: The aim of this study was to compare the correlations between computed tomography (CT) criteria for hepatic steatosis and lipid profile values when hepatic steatosis is incidentally detected. PARTICIPANTS AND METHODS: This is an institutional Review Board-approved, HIPPA-compliant, retrospective study of abdominal CT scans in 200 randomly selected patients who had either nonenhanced CT (NECT) or contrast-enhanced CT (CECT) studies with reported fatty liver. The participants were matched for age, sex, and ethnicity with 200 patients with nonfatty liver. For NECT, four different criteria have been proposed in the literature to diagnose fatty liver: (i) liver Hounsfield Units (HU) less than 48 HU, (ii) ratio of liver to spleen HU less than 0.8, (iii) HU difference between liver and spleen less than -10, and (iv) hepatic vessel HU greater than or equal to liver HU. For CECT, the criteria was attenuation difference between liver and spleen HU, in the portal venous phase of up to -20 to -25 HU. Laboratory results (low-density lipoprotein, high-density lipoprotein, triglycerides) were documented. Matched analyses and conditional logistic regression analysis were carried out for matched variables. RESULTS: There were statistically significant differences in triglyceride values, between the cases and controls (P=0.02), when all criteria were considered. Also, statistically significant differences were found between cases and controls on the basis of NECT criterion 2 and high-density lipoprotein (P=0.04), as well as CECT criteria and triglyceride levels (P=0.02). In addition, the data indicate that criteria for steatosis on CECT may be more broad than traditionally utilized. CONCLUSION: Incidental reporting of fatty liver on NECT/CECT should prompt consideration of clinical follow-up and lipid profile testing in an otherwise asymptomatic patient. Additional metrics for the diagnosis of steatosis in CECT exam should also be considered.

How one institution overcame the challenges to start an MRI-based brachytherapy program for cervical cancer.

PURPOSE: Adaptive magnetic resonance imaging (MRI)-based brachytherapy results in improved local control and decreased high-grade toxicities compared to historical controls. Incorporating MRI into the workflow of a department can be a major challenge when initiating an MRI-based brachytherapy program. This project aims to describe the goals, challenges, and solutions when initiating an MRI-based cervical cancer brachytherapy program at our institution. MATERIAL AND METHODS: We describe the 6-month multi-disciplinary planning phase to initiate an MRI-based brachytherapy program. We describe the specific challenges that were encountered prior to treating our first patient. RESULTS: We describe the solutions that were realized and executed to solve the challenges that we faced to establish our MRI-based brachytherapy program. We emphasize detailed coordination of care, planning, and communication to make the workflow feasible. We detail the imaging and radiation physics solutions to safely deliver MRI-based brachytherapy. The focus of these efforts is always on the delivery of optimal, state of the art patient care and treatment delivery within the context of our available institutional resources. CONCLUSIONS: Previous publications have supported a transition to MRI-based brachytherapy, and this can be safely and efficiently accomplished as described in this manuscript.

Utility of magnetic resonance imaging in establishing a venous pressure gradient in a patient with possible nutcracker syndrome.

Nutcracker syndrome is characterized by abnormal acute angulation of the superior mesenteric artery origin from the aorta, with resulting compression and hypertension of the crossing left renal vein. The radiologic studies used in diagnosis are typically limited to standard cross-sectional anatomic imaging with computed tomography or magnetic resonance imaging, with occasional use of Doppler ultrasound imaging for hemodynamic quantification. The standard for acquiring anatomic and physiologic information continues to be invasive venography. We describe the successful novel application of phase-encoded magnetic resonance imaging as a noninvasive method for acquiring anatomic and hemodynamic data in a case of possible nutcracker syndrome in a young patient.