Faculty Bibliography

PURPOSE/OBJECTIVE:With clinical volumes decreased, radiologists volunteered to participate virtually in daily clinical rounds and provide communication between frontline physicians and patients with coronavirus disease 2019 (COVID-19) and their families affected by restrictive hospital visitation policies. The purpose of this survey-based assessment was to demonstrate the beneficial effects of radiologist engagement during this pandemic and potentially in future crises if needed. METHODS:After the program's completion, a survey consisting of 13 multiple-choice and open-ended questions was distributed to the 69 radiologists who volunteered for a minimum of 7 days. The survey focused on how the experience would change future practice, the nature of interaction with medical students, and the motivation for volunteering. The electronic medical record system identified the patients who tested positive for or were suspected of having COVID-19 and the number of notes documenting family communication. RESULTS:In all, 69 radiologists signed or cosigned 7,027 notes. Of the 69 radiologists, 60 (87.0%) responded to the survey. All found the experience increased their understanding of COVID-19 and its effect on the health care system. Overall, 59.6% agreed that participation would result in future change in communication with patients and their families. Nearly all (98.1%) who worked with medical students agreed that their experience with medical students was rewarding. A majority (82.7%) chose to participate as a way to provide service to the patient population. CONCLUSION/CONCLUSIONS:This program provided support to frontline inpatient teams while also positively affecting the radiologist participants. If a similar situation arises in the future, this communication tool could be redeployed, especially with the collaboration of medical students.

BACKGROUND:The aim of this study was to describe associated injuries in cases of distal biceps femoris avulsions (DBFA) as well as the incidence of neurological injury and radiographic abnormalities of the common peroneal nerve (CPN). METHODS:A retrospective chart review was conducted of patients presenting to our office or trauma center with DBFA injuries. Demographic data was obtained as well as mechanism of injury. Assessment of concomitant injuries and presence of neurologic injury was completed via chart review and magnetic resonance imaging (MRI) review. The CPN was evaluated for signs of displacement or neuritis. RESULTS:Sixteen patients were identified (mean age-at-injury 28.6 years, 87.5% male) with DBFA. Three patients (18.8%) sustained their injuries secondary to high energy trauma while 13 (81.3%) had injuries secondary to lower energy trauma. Nine patients (56.3%) initially presented with CPN palsy. All patients presenting with CPN palsy of any kind were found to have a displaced CPN on MRI and no patient with a normal nerve course had a CPN palsy. CONCLUSIONS:This case series demonstrates a strong association between DBFA and CPN palsy as well as multi-ligamentous knee injury (MLKI). These injuries have a higher rate of CPN palsy than that typically reported for MLKI. Furthermore, these findings suggest that CPN displacement on MRI may be a clinically significant indicator of nerve injury. LOE: IV.

OBJECTIVE:Deep Learning (DL) image reconstruction has the potential to disrupt the current state of MR imaging by significantly decreasing the time required for MR exams. Our goal was to use DL to accelerate MR imaging in order to allow a 5-minute comprehensive examination of the knee, without compromising image quality or diagnostic accuracy. METHODS:A DL model for image reconstruction using a variational network was optimized. The model was trained using dedicated multi-sequence training, in which a single reconstruction model was trained with data from multiple sequences with different contrast and orientations. Following training, data from 108 patients were retrospectively undersampled in a manner that would correspond with a net 3.49-fold acceleration of fully-sampled data acquisition and 1.88-fold acceleration compared to our standard two-fold accelerated parallel acquisition. An interchangeability study was performed, in which the ability of 6 readers to detect internal derangement of the knee was compared for the clinical and DL-accelerated images. RESULTS:The study demonstrated a high degree of interchangeability between standard and DL-accelerated images. In particular, results showed that interchanging the sequences would result in discordant clinical opinions no more than 4% of the time for any feature evaluated. Moreover, the accelerated sequence was judged by all six readers to have better quality than the clinical sequence. CONCLUSIONS:An optimized DL model allowed for acceleration of knee images which performed interchangeably with standard images for the detection of internal derangement of the knee. Importantly, readers preferred the quality of accelerated images to that of standard clinical images.

Purpose/UNASSIGNED:To use convolutional neural networks (CNNs) for fully automated MRI segmentation of the glenohumeral joint and evaluate the accuracy of three-dimensional (3D) MRI models created with this method. Materials and Methods/UNASSIGNED:Shoulder MR images of 100 patients (average age, 44 years; range, 14-80 years; 60 men) were retrospectively collected from September 2013 to August 2018. CNNs were used to develop a fully automated segmentation model for proton density-weighted images. Shoulder MR images from an additional 50 patients (mean age, 33 years; range, 16-65 years; 35 men) were retrospectively collected from May 2014 to April 2019 to create 3D MRI glenohumeral models by transfer learning using Dixon-based sequences. Two musculoskeletal radiologists performed measurements on fully and semiautomated segmented 3D MRI models to assess glenohumeral anatomy, glenoid bone loss (GBL), and their impact on treatment selection. Performance of the CNNs was evaluated using Dice similarity coefficient (DSC), sensitivity, precision, and surface-based distance measurements. Measurements were compared using matched-pairs Wilcoxon signed rank test. Results/UNASSIGNED:value range, .097-.99). Conclusion/UNASSIGNED:© RSNA, 2020.

OBJECTIVE:To report radiographic features and complications of coracoclavicular ligament reconstruction and the association of radiographic features with symptomatology. MATERIALS AND METHODS/METHODS:Retrospective picture archiving and communication system query (1/2012-8/2018) identified subjects with prior coracoclavicular ligament reconstruction. Post-operative radiographs were reviewed with attention to the following: (1) acromioclavicular alignment, (2) coracoclavicular width, (3) distal clavicular osteolysis, (4) osseous tunnel widening, and (5) hardware complication or fracture. Medical records were reviewed to determine purpose of imaging follow-up (symptomatic versus routine). Statistical analysis determined associations between binary features and outcomes, and inter-reader agreement. RESULT/RESULTS:Review of 55 charts identified 32 subjects (23 male, 9 females; age range 24-64; imaged 1-34 months following surgery) meeting inclusion criteria. Loss of acromioclavicular reduction was the most common imaging finding (n = 25, 78%), with 76% progressing to coracoclavicular interval widening. Distal clavicular osteolysis was seen in 21 cases (66%) and was significantly associated with loss of acromioclavicular joint reduction (p = 0.032). Tunnel widening occurred in 23 patients (82%) with more than one follow-up radiograph. Six (19%) had hardware complication or fracture. No radiographic feature or complication had significant correlation with symptomatology (p values 0.071-0.721). Inter-reader agreement was moderate to substantial for coracoclavicular interval widening and hardware complication, fair to substantial for tunnel widening, and fair to moderate for loss of acromioclavicular reduction and distal clavicular osteolysis. CONCLUSION/CONCLUSIONS:Loss of acromioclavicular joint reduction, coracoclavicular interval widening, distal clavicular osteolysis, and tunnel widening are common radiographic features after coracoclavicular ligament reconstruction; however, they do not necessarily correlate with symptomatology.

MR imaging is the standard diagnostic modality that provides a comprehensive and accurate assessment for both osseous and soft-tissue pathologic conditions of the shoulder. This article discusses standard MR imaging and arthrography protocols used routinely in clinical practice, as well as more innovative sequences and reconstruction techniques, facilitated by the increasing availability of high-field-strength magnets and multichannel phased array surface coils and incorporation of artificial intelligence. These exciting innovations allow for a more detailed and diagnostic imaging assessment, improvements in image quality, and more rapid image acquisition.

The focus of this article is to illustrate various pathologic entities and variants, heralding disease about the ankle, based on scrutiny of AP radiographs of the ankle, with correlative findings on cross-sectional imaging. Many of these entities can only be detected on the AP ankle radiograph and, if not recognized, may lead to delayed diagnosis and persistent morbidity to the patient. However, a vigilant radiologist, equipped with the knowledge of the characteristic appearance and typical locations of the imaging findings, should be able to make the crucial initial diagnosis and surmise additional findings to be confirmed on cross-sectional imaging.

Background/Purpose : Prevalence of sacroiliitis (SI) in Crohn's disease (CD) varies widely (range 4% -39%), depending on criteria utilized to define the disease (e.g. inflammatory back pain, plain radiographs or MRI). Sacroiliitis may remain underdiagnosed in CD patients given lack of association with clinical symptoms of back pain and CD activity. However, patients with CD often undergo magnetic resonance enterography (MRE) to assess extent, severity of small bowel CD and radiographic healing, affording clinicians the opportunity to evaluate for the presence of active and/or chronic SI. We sought to identify the prevalence of sacroiliitis in CD patients utilizing MRE and determine its relationship with CD activity, especially with concurrent biologic therapy. Methods : All CD subjects undergoing MRE between years 2014-2018 at a large IBD referral center were identified. A musculoskeletal radiologist, blinded to clinical data, reviewed all MRE exams for the presence of acute bone marrow edema (BME) lesions and chronic lesions suggestive of acute and chronic SI, respectively. A second radiologist, also blinded, assessed MRE for mucosal CD activity using validated measures. Charts were reviewed for demographics, IBD characteristics, presence of back pain, clinical and endoscopic activity of CD, and Crohn's therapies within 3 months of MRE. Comparisons were made between CD subjects with and without SI using chi-square test. Univariate and multivariate logistic regression were used to determine risk factors of SI. Results : 258 subjects with CD underwent MRE during the study period with a mean age of 35 years old, 53% (n=138) were male, and mean duration of CD at the time of MRE was 9 years. Few reported back pain (8%) and 14% had previously seen a rheumatologist. Overall, 17% (n=45) of patients had MR evidence of sacroiliitis (Table 1). Female gender, presence of back pain, and later age of CD diagnosis were associated with signs of sacroiliitis (p=0.05, p< 0.001, p=0.04 respectively; Table 2). Stricturing phenotype was associated with a lower rate of SI (7% vs. 24%; p=0.018), but inflammatory or penetrating phenotypes were not. CD location, activity as noted by clinical scores, endoscopic disease activity, or radiographic disease activity on MRE, were not associated with sacroiliitis (Table 2). On multivariable analysis, back pain was associated with the presence of sacroiliitis on MRE (OR 3.0, 95% CI 1.1-5.6; p=0.04). Concurrent CD therapy with biologics did not lower the risk of sacroiliitis. Conclusion : Although often underdiagnosed, SI is a common comorbid condition in CD. While recent history of back pain was associated with the presence of sacroiliitis visualized on MRE, no correlations were found with other clinical and endoscopic markers of CD activity. Moreover, concurrent CD therapy, especially biologics, was not associated with a lower risk of sacroiliitis on MRE. With limited clinical clues and CD characteristics to suggest sacroiliitis, gastroenterologists can utilize MRE as a screening tool to detect SI and refer CD patients to rheumatologists. Presence of SI on MRE in CD patients with back pain may help identify a subset of individuals likely to benefit from switching to therapies with proven efficacy in axial SpA