Publications

BACKGROUND: Thresholding apparent diffusion coefficient (ADC) maps obtained from Diffusion-Weighted-Imaging (DWI) has been proposed for identifying benign lesions that can safely avoid biopsy. The presence of malignancies with high ADC values leads to high thresholds, limiting numbers of avoidable biopsies.

PURPOSE: We evaluate two previously reported methods for identifying avoidable biopsies: using case-set dependent ADC thresholds that assure 100% sensitivity and using negative likelihood ratio (LR-) with a fixed ADC threshold of 1.50 × 10-3 mm2/s. We evaluated improvements in efficacy obtained by excluding non-mass lesions and lesions with anisotropic intra-lesion morphologic characteristics.

STUDY TYPE: Prospective.

POPULATION: 55 adult females with dense breasts with 69 BI-RADS 4 or 5 lesions (38 malignant, 31 benign) identified on ultrasound and mammography and imaged with MRI prior to biopsy.

FIELD STRENGTH/SEQUENCE: 1.5 T and 3.0 T. DWI.

ASSESSMENT: Analysis of DWI, including directional images was done on an ROI basis. ROIs were drawn on DWI images acquired prior to biopsy, referencing all available images including DCE, and mean ADC was measured. Anisotropy was quantified via variation in ADC values in the lesion core across directional DWI images.

STATISTICAL TESTS: Improvement in specificity at 100% sensitivity was evaluated with exact McNemar test with 1-sided p-value < 0.05 indicating statistical significance.

RESULTS: Using ADC thresholding that assures 100% sensitivity, non-mass and directional variance filtering improved the percent of avoidable biopsies to 42% from baseline of 10% achieved with ADC thresholding alone. Using LR-, filtering improved outcome to 0.06 from baseline 0.25 with ADC thresholding alone. ADC thresholding showed a lower percentage of avoidable biopsies in our cohort than reported in prior studies. When ADC thresholding was supplemented with filtering, the percentage of avoidable biopsies exceeded those of prior studies.

DATA CONCLUSION: Supplementing ADC thresholding with filters excluding non-mass lesions and lesions with anisotropic characteristics on DWI can result in an increased number of avoidable biopsies.

Arterial spin labeling (ASL) is a non-invasive MRI technique that allows for quantitative measurement of cerebral perfusion. Incomplete or inaccurate reporting of acquisition parameters complicates quantification, analysis, and sharing of ASL data, particularly for studies across multiple sites, platforms, and ASL methods. There is a strong need for standardization of ASL data storage, including acquisition metadata. Recently, ASL-BIDS, the BIDS extension for ASL, was developed and released in BIDS 1.5.0. This manuscript provides an overview of the development and design choices of this first ASL-BIDS extension, which is mainly aimed at clinical ASL applications. Discussed are the structure of the ASL data, focussing on storage order of the ASL time series and implementation of calibration approaches, unit scaling, ASL-related BIDS fields, and storage of the labeling plane information. Additionally, an overview of ASL-BIDS compatible conversion and ASL analysis software and ASL example datasets in BIDS format is provided. We anticipate that large-scale adoption of ASL-BIDS will improve the reproducibility of ASL research.

BACKGROUND: Patients with interatrial shunts (patient foramen ovale/atrial septal defect) are potentially at increased risk for paradoxical air embolism following computed tomography (CT) scans with intravenous (IV) contrast media injection. IV in-line filters aim to prevent such embolisms but are not compatible with power injection required for diagnostic CT.

PURPOSE: The purpose of this study was to determine whether the incidence of paradoxical embolism to the heart and brain in patients with an interatrial shunt is higher compared with controls within 48 hours following injection of IV contrast media without IV in-line filter.

METHODS: This is a retrospective cohort study conducted at a large tertiary academic center, which included a total of 2929 consecutive patients who underwent 8983 CT scans with IV contrast media injection between July 1, 2000 and April 30, 2018. Diagnosis of an interatrial shunt was confirmed by transthoracic or transesophageal echocardiography. Incidence and risk of cardiac embolic events (new troponin elevation, >0.1 ng/mL) and neurological embolic events (new diagnosis of stroke/transient ischemic attacks) were evaluated.

RESULTS: Among the 2929 patients analyzed (mean±SD age, 61±14 y), 475/2929 (16.2%) patients had an interatrial shunt. After applying the exclusion criteria, new elevated troponin was found in 8/329 (2.4%; 95% confidence interval [CI]: 1.1-4.7) patients with an interatrial shunt compared with 25/1687 (1.5%; 95% CI: 0.9-2.2) patients without an interatrial shunt. New diagnosis of stroke occurred in 2/169 (1%; 95% CI: 0.3-4.2) of patients with an interatrial shunt compared with 7/870 (0.8%; 95% CI: 0.4-1.7) without interatrial shunt.

CONCLUSION: Among patients with echocardiographic evidence of an interatrial shunt, IV CT contrast administration without an in-line filter does not increase the incidence of cardiac or neurological events.

BACKGROUND: Four-dimensional (4D) CT localization allows minimally invasive parathyroidectomy as treatment for primary hyperparathyroidism (PHPT), but false positive localization is frequent. We sought to characterize the ability of 4D CT to predict four-gland hyperplasia (HP) based on the size of candidate lesions.

STUDY DESIGN: We retrospectively analyzed patients with PHPT who underwent 4D CT imaging and parathyroidectomy between 2014 and 2020 from a prospectively collected institutional database. The cohort was stratified into two groups, HP vs single adenoma (SA) and double adenoma (DA), based on operative findings and pathology. Logistic regression models assessed the association between the greatest diameter of the dominant candidate lesion on 4D CT and the outcomes of four-gland hyperplasia vs SA and DA.

RESULTS: Among a cohort of 240 patients, 41 were found to have HP, and 199 had adenomas (SA = 155, DA = 44). Patients with HP were less likely to have a preoperative calcium level greater than 1 mg/dL above the upper limit of normal compared with patients with adenomas (63% vs 81%, p = 0.02) and more likely to report symptoms (61% vs 43%, p = 0.04). After adjusting for BMI, we found an estimated 13% reduction in odds of HP for every 1-mm increase in the greatest diameter of dominant candidate lesions identified on 4D CT scan (odds ratio 0.87, 95% CI 0.78 to 0.96, p = 0.009).

CONCLUSIONS: A smaller size of the dominant lesion on 4D CT scan is associated with an increased risk of HP in PHPT. Use of 4D CT imaging localization may provide evidence for differentiating HP from adenomas.

BACKGROUND. A standardized guideline and scoring system would improve evaluation and reporting of peripheral neuropathy (PN) on MRI. OBJECTIVE. The objective of this study was to create and validate a neuropathy classification and grading system, which we named the Neuropathy Score Reporting and Data System (NS-RADS). METHODS. This retrospective study included 100 patients with nerve imaging studies and known clinical diagnoses. Experts crafted NS-RADS using mutually agreed-on qualitative criteria for the classification and grading of PN. Different classes were created to account for the spectrum of underlying pathologies: unremarkable (U), injury (I), neoplasia (N), entrapment (E), diffuse neuropathy (D), not otherwise specified (NOS), and postintervention state (PI). Subclasses were established to describe the severity or extent of the lesions. Validation testing was performed by 11 readers from 10 institutions with experience levels ranging from 3 to 18 years after residency. After initial reader training, cases were presented to readers who were blinded to the final clinical diagnoses. Interobserver agreement was assessed using correlation coefficients and the Conger kappa, and accuracy testing was performed. RESULTS. Final clinical diagnoses included normal (n = 5), nerve injury (n = 25), entrapment (n = 15), neoplasia (n = 33), diffuse neuropathy (n = 18), and persistent neuropathy after intervention (n = 4). The miscategorization rate for NS-RADS classes was 1.8%. Final diagnoses were correctly identified by readers in 71-88% of cases. Excellent inter-reader agreement was found on the NS-RADS pathology categorization (κ = 0.96; 95% CI, 0.93-0.98) as well as muscle pathology categorization (κ = 0.76; 95% CI, 0.68-0.82). The accuracy for determining milder versus more severe categories per radiologist ranged from 88% to 97% for nerve lesions and from 86% to 94% for muscle abnormalities. CONCLUSION. The proposed NS-RADS classification is accurate and reliable across different reader experience levels and a spectrum of PN conditions. CLINICAL IMPACT. NS-RADS can be used as a standardized guideline for reporting PN and improved multidisciplinary communications.

PURPOSE: We constructed a 13C/31P surface coil at 3 T for studying cancer metabolism and bioenergetics. In a single scan session, hyperpolarized 13C-pyruvate MRS and 31P MRS was carried out for a healthy rat brain.

METHODS: All experiments were carried out at 3 Tesla. The multinuclear surface coil was designed as two coplanar loops each tuned to either the 13C or 31P operating frequency with an LCC trap on the 13C loop. A commercial volume proton coil was used for anatomical localization and B0 shimming. Single tuned coils operating at either the 13C or 31P frequency were built to evaluate the relative performance of the multinuclear coil. Coil performance metrics consisted of measuring Q factor ratio, calculating system input power using a single-pulse acquisition, and acquiring SNR and flip angle maps using 2D CSI sequences. To observe in vivo spectra, a bolus of hyperpolarized [1-13C] pyruvate was administered via tail vein. In vivo13C and endogenous 31P spectra were obtained in a single scan session using 1D slice selective acquisitions.

RESULTS: When compared with single tuned surface coils, the multinuclear coil performance showed a decrease in Q factor ratio, SNR, and transmit efficiency. Flip angle maps showed adequate flip angles within the phantom when the transmit voltage was set using an external phantom. Results show good detection of 13C labeled lactate, alanine, and bicarbonate in addition to ATP from 31P MRS.

CONCLUSIONS: The coil enables obtaining complementary information within a scan session, thus reducing the number of trials and minimizing biological variability for studies of metabolism and bioenergetics.

BACKGROUND: Exercise cardiovascular magnetic resonance (Ex-CMR) is a promising stress imaging test for coronary artery disease (CAD). However, Ex-CMR requires accelerated imaging techniques that result in significant aliasing artifacts. Our goal was to develop and evaluate a free-breathing and electrocardiogram (ECG)-free real-time cine with deep learning (DL)-based radial acceleration for Ex-CMR.

METHODS: A 3D (2D + time) convolutional neural network was implemented to suppress artifacts from aliased radial cine images. The network was trained using synthetic real-time radial cine images simulated using breath-hold, ECG-gated segmented Cartesian k-space data acquired at 3 T from 503 patients at rest. A prototype real-time radial sequence with acceleration rate = 12 was used to collect images with inline DL reconstruction. Performance was evaluated in 8 healthy subjects in whom only rest images were collected. Subsequently, 14 subjects (6 healthy and 8 patients with suspected CAD) were prospectively recruited for an Ex-CMR to evaluate image quality. At rest (n = 22), standard breath-hold ECG-gated Cartesian segmented cine and free-breathing ECG-free real-time radial cine images were acquired. During post-exercise stress (n = 14), only real-time radial cine images were acquired. Three readers evaluated residual artifact level in all collected images on a 4-point Likert scale (1-non-diagnostic, 2-severe, 3-moderate, 4-minimal).

RESULTS: The DL model substantially suppressed artifacts in real-time radial cine images acquired at rest and during post-exercise stress. In real-time images at rest, 89.4% of scores were moderate to minimal. The mean score was 3.3 ± 0.7, representing increased (P < 0.001) artifacts compared to standard cine (3.9 ± 0.3). In real-time images during post-exercise stress, 84.6% of scores were moderate to minimal, and the mean artifact level score was 3.1 ± 0.6. Comparison of left-ventricular (LV) measures derived from standard and real-time cine at rest showed differences in LV end-diastolic volume (3.0 mL [- 11.7, 17.8], P = 0.320) that were not significantly different from zero. Differences in measures of LV end-systolic volume (7.0 mL [- 1.3, 15.3], P < 0.001) and LV ejection fraction (- 5.0% [- 11.1, 1.0], P < 0.001) were significant. Total inline reconstruction time of real-time radial images was 16.6 ms per frame.

CONCLUSIONS: Our proof-of-concept study demonstrated the feasibility of inline real-time cine with DL-based radial acceleration for Ex-CMR.

INTRODUCTION: Morphometric assessment of Chiari malformation type I (CMI) is typically performed on a midsagittal MRI. However, errors arising from an imprecise selection of the midsagittal plane are unknown. We define absolute parasagittal error as the absolute difference between morphometric measurements at the midsagittal and parasagittal planes. Our objective was to determine the absolute parasagittal error at various lateral distances for morphometric parameters commonly used in CMI research.

METHODS: Sagittal T1-weighted MRI scans of 30 CMI adult female subjects were included. Image sets were evaluated to assess 14 CMI morphometric parameters in the midsagittal plane and four parasagittal planes located 1 and 2 mm lateral (left and right). Comparisons between measurements at the midsagittal and parasagittal planes were conducted to determine the mean individual absolute and mean group parasagittal errors for all 14 parameters.

RESULTS: The mean individual absolute parasagittal error was > 1 unit (1 mm for lengths and 1 degree for angles) for 9/14 parameters within a lateral distance of 2 mm. No significant group parasagittal errors were seen in 14/14 parameters, including tonsillar position within a lateral distance of 2 mm.

CONCLUSION: Our results show that the absolute errors for imprecise midsagittal plane selection may impact the clinical assessment of an individual patient. However, the impact on group measurements, such as in a research setting, will be minimal.

RATIONALE AND OBJECTIVES: In response to COVID-19, our institution implemented three virtual readout systems: a commercial HIPAA compliant web-based video conferencing platform used for screen-sharing (Starleaf), an interactive control sharing system integrated into PACS allowing simultaneous multi-user mouse control over images (Collaborate), and the telephone. Our aim was to assess overall satisfaction with and perceived effectiveness of these virtual readout methods to optimize best practices for the future.

MATERIALS AND METHODS: An IRB-exempt survey was electronically distributed to 64 trainees and 76 attendings at one tertiary-care institution via Survey Monkey. Questions focused on overall satisfaction, perceived effectiveness, technical difficulties, and continued future use of the three virtual readout strategies. Answers were collected with Likert scales, tick boxes, and open-ended questions.

RESULTS: 32/64 trainees (50%) and 32/76 attendings (42%) completed the survey. Trainees and attendings were more satisfied with screen sharing (Starleaf) and perceived it more effective than control sharing (Collaborate) or the telephone (p < 0.0001). Respondents experienced more technical difficulties with control sharing versus screen sharing (p = 0.0004) with a negative correlation between level of technical difficulties and satisfaction with screen sharing (r = -0.50, p < 0.0001) and control sharing (r = -0.38, p = 0.0006). Trainees and faculty supported a combination of in-person and virtual readouts in the future (p < 0.0001).

CONCLUSION: Platforms mirroring in-person readouts, such as Starleaf, are preferred by both trainees and attendings over non-screen sharing platforms such as the telephone. However, technical stability determines satisfaction between similar platforms. Both trainees and attendings support incorporation of virtual readout methods in combination with traditional in-person readouts in the post-COVID-19 era.

PURPOSE: To describe the feasibility and outcomes of filter-assisted shunt embolization in patients with acquired large portosystemic shunts.

METHODS: Two-center HIPAA compliant retrospective study of all patients who underwent filter-assisted shunt embolization between 03/2015-03/2021. Initial clinical evaluation, including demographic information, shunt sizing, and procedural details, was reviewed. Technical success was defined as successful occlusion of the targeted shunt.

RESULTS: Eight patients (55 ± 10 years/88% male) had a large acquired portosystemic shunt which was suitable for filter-assisted shunt embolization. Indications for the procedure: 3 = pre-transplantation optimization, 2 = overt hepatic encephalopathy (HE), 1 = post-transplant thrombosis, 1 = portal vein thrombosis and encephalopathy, 1 = encephalopathy and variceal bleeding. Portosystemic shunts occurred between splenic and renal veins, inferior mesenteric and gonadal veins, and coronary veins. Mean shunt diameter was 27 ± 5 mm. The technical success of the procedure was 8/8 (100%). In 7 patients, a transjugular intrahepatic portosystemic shunt (TIPS) was also placed at the time of the shunt embolization due to either pre-transplant indication or sluggish portal flow. There were no intraprocedural complications from filter placement.

OUTCOMES: 3 = currently listed for transplant, 2 = resolution of HE, 1 = made CMO, 1 patient with patent post-transplant vasculature. 1 = died as a complication related to TIPS placement (SIR Class F Complication).

CONCLUSION: Filter-assisted shunt embolization is a technically feasible and safe technique to reduce or embolize large portosystemic shunts.