Integrating Eye Tracking and Speech Recognition Accurately Annotates MR Brain Images for Deep Learning: Proof of Principle.

PURPOSE: To generate and assess an algorithm combining eye tracking and speech recognition to extract brain lesion location labels automatically for deep learning (DL). MATERIALS AND METHODS: In this retrospective study, 700 two-dimensional brain tumor MRI scans from the Brain Tumor Segmentation database were clinically interpreted. For each image, a single radiologist dictated a standard phrase describing the lesion into a microphone, simulating clinical interpretation. Eye-tracking data were recorded simultaneously. Using speech recognition, gaze points corresponding to each lesion were obtained. Lesion locations were used to train a keypoint detection convolutional neural network to find new lesions. A network was trained to localize lesions for an independent test set of 85 images. The statistical measure to evaluate our method was percent accuracy. RESULTS: Eye tracking with speech recognition was 92% accurate in labeling lesion locations from the training dataset, thereby demonstrating that fully simulated interpretation can yield reliable tumor location labels. These labels became those that were used to train the DL network. The detection network trained on these labels predicted lesion location of a separate testing set with 85% accuracy. CONCLUSION: The DL network was able to locate brain tumors on the basis of training data that were labeled automatically from simulated clinical image interpretation.© RSNA, 2020.

Optimisation in daily practice - it's more than just radiation dose.

On this editorial, the importance of optimisation of diagnostic imaging and radiation therapy in terms of diagnostic and therapeutic tests and treatments, education and service provision are discussed.

The State of Radiology AI: Considerations for Purchase Decisions and Current Market Offerings.

PURPOSE: To provide an overview of important factors to consider when purchasing radiology artificial intelligence (AI) software and current software offerings by type, subspecialty, and modality. MATERIALS AND METHODS: Important factors for consideration when purchasing AI software, including key decision makers, data ownership and privacy, cost structures, performance indicators, and potential return on investment are described. For the market overview, a list of radiology AI companies was aggregated from the Radiological Society of North America and the Society for Imaging Informatics in Medicine conferences (November 2016-June 2019), then narrowed to companies using deep learning for imaging analysis and diagnosis. Software created for image enhancement, reporting, or workflow management was excluded. Software was categorized by task (repetitive, quantitative, explorative, and diagnostic), modality, and subspecialty. RESULTS: A total of 119 software offerings from 55 companies were identified. There were 46 algorithms that currently have Food and Drug Administration and/or Conformité Européenne approval (as of November 2019). Of the 119 offerings, distribution of software targets was 34 of 70 (49%), 21 of 70 (30%), 14 of 70 (20%), and one of 70 (1%) for diagnostic, quantitative, repetitive, and explorative tasks, respectively. A plurality of companies are focused on nodule detection at chest CT and two-dimensional mammography. There is very little activity in certain subspecialties, including pediatrics and nuclear medicine. A comprehensive table is available on the website hitilab.org/pages/ai-companies. CONCLUSION: The radiology AI marketplace is rapidly maturing, with an increase in product offerings. Radiologists and practice administrators should educate themselves on current product offerings and important factors to consider before purchase and implementation.© RSNA, 2020See also the invited commentary by Sala and Ursprung in this issue.

Daily Caffeine Consumption Is Associated with Decreased Incidence of Symptoms and Hemodynamic Changes During Pharmacologic Stress with Regadenoson.

Regadenoson is an adenosine A2A receptor agonist widely used as a pharmacologic stress agent for myocardial perfusion imaging. Approximately 3.4 million regadenoson pharmacologic stress tests were performed annually as of 2011. Caffeine is a competitive antagonist of all adenosine receptor subtypes; thus, caffeine is typically withheld 12-24 h before stress with regadenoson. However, the effects of daily caffeine intake on regadenoson stress are unknown. This study assessed the effects of daily caffeine intake on symptoms and hemodynamic changes during stress testing with regadenoson. Methods: Patients presenting for regadenoson stress myocardial perfusion imaging were asked their amounts of daily caffeine intake. Chart review was used to collect data on demographics, comorbidities, and use of ß-blockers. Data collected from the regadenoson stress test included symptoms, administration of aminophylline, heart rate, blood pressure, and arrhythmias. χ2 testing and ANOVA were used to analyze data divided into 3 categories of caffeine intake (<200, 200-400, and >400 mg daily). χ2 testing was used for nominal data, and unpaired t testing was used for continuous data. Results: In total, 101 patients were enrolled: 53% men and 47% women. Of the 101 patients, 89% reported caffeine intake, with 13% reporting heavy caffeine intake (>400 mg daily). The last intake of caffeine was at least 12 h before the test. During the test, 63% of patients reported symptoms, but the test was completed successfully in all patients. Compared with those who do not use caffeine, intake for caffeine users was associated with less chest pain (P = 0.0013), less aminophylline administration (P = 0.0371), lower resting and peak heart rate (P = 0.0497 and 0.0314, respectively), and lower diastolic blood pressure response (P = 0.0468). No associations were found between caffeine intake and arrhythmia or systolic blood pressure response. Conclusion: The use of regadenoson stress for myocardial perfusion imaging in caffeine consumers is very common, safe, and associated with a lower incidence of certain symptoms than in non-caffeine consumers. Specifically, caffeine intake was associated with less aminophylline use and chest pain.

Invention and Early History of Telepathology (1985-2000).

This narrative-based paper provides a first-person account of the early history of telepathology (1985-2000) by the field's inventor, Ronald S. Weinstein, M. D. During the 1980s, Dr. Weinstein, a Massachusetts General Hospital-trained pathologist, was director of the Central Pathology Laboratory (CPL) for the National Cancer Institute-funded National Bladder Cancer Project, located at Rush Medical College in Chicago, IL. The CPL did post therapy revalidations of surgical pathology and cytopathology diagnoses before outcomes of the completed clinical trials were published. The CPL reported that interobserver variability was invalidating inclusion of dozens of treated bladder cancer patients in published reports on treatment outcomes. This problem seemed ripe for a technology-assisted solution. In an effort to solve the interobserver variability problem, Dr. Weinstein devised a novel solution, dynamic-robotic telepathology, that would potentially enable CPL uropathologists to consult on distant uropathology cases in real-time before their assignment to urinary bladder cancer, tumor stage, and grade-specific clinical trials. During the same period, universities were ramping up their support for faculty entrepreneurism and creating in-house technology transfer organizations. Dr. Weinstein recognized telepathology as a potential growth industry. He and his sister, Beth Newburger, were a successful brother-sister entrepreneur team. Their PC-based education software business, OWLCAT™, had just been acquired by Digital Research Inc., a leading software company, located in California. With funding from the COMSAT Corporation, a publically traded satellite communications company, the Weinstein-Newburger team brought the earliest dynamic-robotic telepathology systems to market. Dynamic-robotic telepathology became a dominant telepathology technology in the late 1990s. Dr. Weinstein, a serial entrepreneur, continued to innovate and, with a team of optical scientists at The University of Arizona's College of Optical Sciences, developed the first sub-1-min whole-slide imaging system, the DMetrix DX-40 scanner, in the early 2000s.

Diffusion Tensor Imaging of the Ankle as a Possible Predictor of Chemotherapy Induced Peripheral Neuropathy: Pilot Study.

PURPOSE: Chemotherapy induced peripheral neuropathy (CIPN) is seen in up to 75% of treated cancer patients and can drastically limit their medical management and affect quality of life. Clinical and electrodiagnostic testing for CIPN have many pitfalls. Magnetic resonance neurography (MRN) is being increasingly used in the evaluation of peripheral nerves. Diffusion tensor imaging (DTI) shows promise in the workup of peripheral nerves. In this prospective pilot study, we investigated a possible relationship between DTI and peripheral neuropathy of the ankle and foot in cancer patients treated with chemotherapy. METHODS: Nine cancer patients with and without CIPN were clinically evaluated using vibratory perception threshold (VPT) testing. VPT score of >25Volts defined presence of CIPN. The posterior tibial nerve and branches in both feet were imaged using MRN and DTI. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) values were measured at the posterior tibial, medial plantar, and lateral plantar nerves. Measurements for the CIPN group were compared to without CIPN by VPT cutoff. Correlations and possible relationships between DTI parameters and CIPN were analyzed. RESULTS: A total of 16feet of 9 enrolled patients were imaged (9feet with CIPN and 7feet without CIPN). Average age was 60.6 ± 13.4 years (range: 33-74). Posterior tibial nerve ADC values were significantly lower than the medial plantar nerve ADC values in all feet (F = 3.50, P = 0.04). We found a correlation with FA and ADC values at specific nerve locations with CIPN, with the left medial plantar nerve FA value and left lateral plantar nerve ADC value demonstrating the strongest positive correlations (0.73 and 0.62, respectively). CONCLUSIONS: The use of DTI for assessing CIPN is challenging but promising. This pilot study provides preliminary data showing correlations between FA and ADC measurements with CIPN and potential utility of DTI as a predictive marker of onset and severity of CIPN in the ankle and foot, which could aid in preventive strategies. Larger, prospective DTI studies are needed to draw definitive conclusions. CLINICAL RELEVANCE: MRN with DTI shows promising results as a potential predictive marker of CIPN in the ankle and foot.

Patient Knowledge Regarding Colorectal Cancer Risk, Opinion of Screening, and Preferences for a Screening Test.

AIMS: To assess patient knowledge about colorectal cancer incidence and prognosis as well as willingness to undergo screening with various tests (eg, optical colonoscopy, stool-based tests, computed tomographic colonography (CTC)). MATERIALS AND METHODS: A survey was administered to consecutive patients of a general academic-based internal medicine clinic. RESULTS: Survey response rate was 86.3%. A majority of respondents (55%) reported being aware of general information about colorectal cancer, and 99% indicated a belief that colorectal cancer screening was a good idea. A majority of respondents (73%) were willing to undergo optical colonoscopy, and some were willing to undergo stool-based tests (48%), or CT colonography CTC (40%). A majority reported being more willing to undergo a colorectal cancer screening test if the test did not involve radiation (86%), did not involve insertion of a tube or device into the rectum (78%), did not involve a pre-proceduralpreprocedural bowel cleansing regimen (73%), or did not involve sedation (60%). CONCLUSION: Improved patient education about the negligible radiation risk associated with CTC or development of a non-invasive imaging test that did not involve a preprocedural bowel cleansing regimen may increase rates of colorectal cancer screening.

Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System.

Purpose To compare breast cancer detection performance of radiologists reading mammographic examinations unaided versus supported by an artificial intelligence (AI) system. Materials and Methods An enriched retrospective, fully crossed, multireader, multicase, HIPAA-compliant study was performed. Screening digital mammographic examinations from 240 women (median age, 62 years; range, 39-89 years) performed between 2013 and 2017 were included. The 240 examinations (100 showing cancers, 40 leading to false-positive recalls, 100 normal) were interpreted by 14 Mammography Quality Standards Act-qualified radiologists, once with and once without AI support. The readers provided a Breast Imaging Reporting and Data System score and probability of malignancy. AI support provided radiologists with interactive decision support (clicking on a breast region yields a local cancer likelihood score), traditional lesion markers for computer-detected abnormalities, and an examination-based cancer likelihood score. The area under the receiver operating characteristic curve (AUC), specificity and sensitivity, and reading time were compared between conditions by using mixed-models analysis dof variance and generalized linear models for multiple repeated measurements. Results On average, the AUC was higher with AI support than with unaided reading (0.89 vs 0.87, respectively; P = .002). Sensitivity increased with AI support (86% [86 of 100] vs 83% [83 of 100]; P = .046), whereas specificity trended toward improvement (79% [111 of 140]) vs 77% [108 of 140]; P = .06). Reading time per case was similar (unaided, 146 seconds; supported by AI, 149 seconds; P = .15). The AUC with the AI system alone was similar to the average AUC of the radiologists (0.89 vs 0.87). Conclusion Radiologists improved their cancer detection at mammography when using an artificial intelligence system for support, without requiring additional reading time. Published under a CC BY 4.0 license. See also the editorial by Bahl in this issue.

The Veil of Obscuration: Additional Radiographic Sign of Posterior Shoulder Dislocation.

OBJECTIVE: To describe a new radiographic sign, "veil of obscuration", associated with posterior glenohumeral joint (shoulder) dislocations and determine its incidence and validity compared to other known classic radiographic signs. METHODS: Four-year retrospective study identified 30 acute posterior shoulder dislocation patients. Radiographs reviewed in consensus by 2 musculoskeletal radiologists for the "veil of obscuration", seen on AP shoulder radiographs and representing a comminuted fracture of the lesser tuberosity projecting over the humeral head or glenohumeral joint. Incidence of this radiographic sign of posterior glenohumeral joint dislocation in addition to other previously described classic radiographic signs, and association with other fractures, surgery, and mechanism of injury were evaluated. Continuous data was analyzed with student t-test and categorical data with Chi-Square test. RESULTS: There were 20 right and 10 left posterior shoulder dislocations. Majority of injuries resulted from vehicle crash (44%). In most cases, reverse Hill-Sachs lesion (83%) and fixed internal rotation of the humeral head (76%) were present, followed by trough line (43%) and "veil of obscuration" (40%). Trough line was seen in significantly more major trauma and vehicle crashes (78% and 46%; P=0.015), while "veil of obscuration" was seen in more seizures (86%; P=0.037) and in all surgical patients. No significant difference in presence of other classic radiographic signs in regards to surgery. CONCLUSION: The newly described radiographic sign of posterior shoulder dislocations named the "veil of obscuration" has comparable incidence as other classic radiographic signs and may be useful in the recognition and diagnosis of these injuries.

A New Software Platform to Improve Multidisciplinary Tumor Board Workflows and User Satisfaction: A Pilot Study.

BACKGROUND: Workflow and preparation for holding multidisciplinary cancer case reviews (i.e., Tumor Boards) is time-consuming and cumbersome. Use of a software platform might improve this process. This pilot study assessed the impact of a new software platform on tumor board preparation workflow and user satisfaction compared to current methods. MATERIALS AND METHODS: Using current methods and the NAVIFY Tumor Board Solution, this study assessed the number of tasks and time to prepare tumor board cases. Participants completed online surveys assessing ease of use and satisfaction with current and new platforms. RESULTS: A total of 41 sessions included two surgeons, two oncologists, two pathologists, and two radiologists preparing tumor board cases with 734 tasks were recorded. Overall, there was no difference in the number of tasks using either preparation method (341 current, 393 NAVIFY Tumor Board solution). There was a significant difference in overall preparation time as a function of specialty (F = 71.74, P < 0.0001), with oncologists, radiologists, and surgeons having reduced times with NAVIFY Tumor Board solution compared to the current platform and pathologists having equivalent times. There was a significant difference (F = 38.98, P < 0.0001) for times as a function of task category. Review of clinical course data and other preparation tasks decreased significantly, but pathology and radiology review did not differ significantly. The new platform received higher ratings than the current methods on all survey questions regarding the ease of use and satisfaction. CONCLUSIONS: The study supported the hypothesis that the new software platform can improve Tumor Board preparation. Further study is needed to assess the impact of this platform in different hospitals, different data storage systems, with different observers, and different types of Tumor board cases as well as its impact on the quality of the tumor board discussion.

Regional Changes in Brain 18F-FDG Uptake After Prophylactic Cranial Irradiation and Chemotherapy in Small Cell Lung Cancer May Reflect Functional Changes.

Chemotherapy followed by prophylactic cranial irradiation (PCI) is associated with increased survival in patients with small cell lung cancer but is associated with fatigue and cognitive impairment. This retrospective study evaluated regional differences in 18F-FDG uptake by the brain before and after PCI. The null hypothesis was that direct toxic effects on the brain from PCI and chemotherapy are symmetric; thus, asymmetric deviations may reflect functional changes due to therapy. Methods: Electronic medical records from 2013 to 2016 were reviewed for patients with small cell lung cancer, MRI of brain negative for metastasis, and 18F-FDG PET/CT scans before and after PCI. As the standard of care, patients received first-line chemotherapy or chemoradiation to the thorax followed by PCI. The 18F-FDG PET/CT scans nearest the PCI were selected. Sixteen patients met these initial criteria. Commercially available PET software was used to register and subtract the PET scans before and after PCI to obtain difference maps. Occipital and cerebellar regions were excluded from the final statistical analysis given the known high variability and misregistration. The χ2 test was used to analyze the data. Results: Two patients had 18F-FDG uptake differences only in the occipital and cerebellar regions. The software registration failed on 1 patient's scans. Therefore, 13 patients were included in the final analysis. Nine of 13 patients demonstrated significant unilateral changes in only 1 region of the brain, and 3 of 13 showed significant changes unilaterally in 2 regions. The χ2 test revealed a significant unilateral regional difference on a patient level (χ2 = 6.24, P = 0.025). The most commonly affected brain region was the frontal lobe. Conclusion: Significantly more patients had unilateral than bilateral regional differences (both increases and decreases) in 18F-FDG uptake in the brain before and after PCI. This finding suggests that differences in unilateral distribution are related to functional changes, since direct toxicity alone from PCI and chemotherapy would be symmetric. The frontal region was the most commonly affected, suggesting a potential contributing etiology for cognitive impairment and decreased executive function after therapy.

Increasing display luminance as a means to enhance interpretation accuracy and efficiency when reducing full-field digital mammography dose.

Reducing dose increases noise impacting image quality but can be offset by increasing display luminance. Two contrast detail mammography images were obtained at 26 kV and the same distance between detectors, at 45 and 50 mAs resulting in entrance surface doses of 7.09 and 7.88 mGy, respectively. They were processed to make average gray level of the background independent of the dose level while maintaining original SNR. Eight radiologists viewed the images at 420, 1000 cd/m2 , and SpotView™ a tool that resulted in an average display luminance of 3138.8 cd/m2 . Percent correct (PC) for all three luminances was higher for high versus low dose. Performance was always higher with high dose no matter what the luminance. For low dose, PC was highest with SpotView™, and SpotView™ and 1000 cd/m2 were significantly higher than 420 cd/m2 . At high dose, SpotView™ PC was significantly higher than both lower luminances. Average time per image was lower in high dose, and, at both doses, time decreased as luminance increased, with SpotView™ having significantly shorter times. Increasing luminance from 420 to 1000 cd/m2 significantly increases target detection by ∼3.0% and with SpotView™ by ∼6.2% . Increasing display luminance with SpotView™ significantly decreases reading time by 16.0%.

18F-FDG PET/CT Can Predict Development of Thyroiditis Due to Immunotherapy for Lung Cancer.

Our primary purpose was to determine whether increased 18F-FDG uptake in the thyroid gland predicts development of thyroiditis with subsequent hypothyroidism in patients undergoing immunotherapy with nivolumab for lung cancer. Secondarily, we determined whether 18F-FDG uptake in the thyroid gland correlates with number of administered cycles of nivolumab. Methods: Retrospective chart review over 2 y found 18 lung cancer patients treated with nivolumab who underwent 18F-FDG PET/CT before and during therapy. SUVmean, SUVmax, and total lesion glycolysis of the thyroid gland were measured. SUVs were also measured for the pituitary gland, liver, and spleen. Patients underwent monthly thyroid testing. PET/CT parameters were analyzed by unpaired t testing for differences between 2 groups (patients who developed hypothyroidism and those who did not). Correlation between development of thyroiditis and number of cycles of nivolumab was also tested. Results: Six of 18 patients developed hypothyroidism. The t test comparing the 2 groups demonstrated significant differences in SUVmean (P = 0.04), SUVmax (P = 0.04), and total lesion glycolysis (P = 0.02) of the thyroid gland. Two of 4 patients who developed thyroiditis and had increased 18F-FDG uptake in the thyroid gland had a normal TSH level at the time of follow-up 18F-FDG PET/CT. Patients who developed thyroiditis with subsequent hypothyroidism stayed longer on therapy (10.6 cycles) than patients without thyroiditis (7.6 cycles), but the trend was not statistically significant. No significant difference in PET/CT parameters was observed for pituitary gland, liver, or spleen. Conclusion:18F-FDG PET/CT can predict the development of thyroiditis with subsequent hypothyroidism before laboratory testing. Further study is required to confirm the positive trend between thyroiditis and duration of therapy.

Addressing Racial Disparity in Colorectal Cancer Screening With CT Colonography: Experience in an African-American Cohort.

BACKGROUND: Incidence and mortality from colorectal cancer (CRC) are higher among African Americans where CRC screening rates lag behind non-Hispanic whites. The purpose of this investigation was to evaluate CT colonography (CTC) performance and outcome measures in an African-American cohort. MATERIALS AND METHODS: This study is an institutional review board-approved and Health Insurance Portability and Accountability Act-compliant study. Retrospective evaluation of the African-American cohort from the Atlanta VA Medical Center (Decatur, Georgia) CTC screening program. Colonoscopy referral rates and histologic diagnoses were evaluated. RESULTS: CTC studies (n = 2490; mean age, 59.5 years [range, 50-80 years]; 85.5% male; all African American) were performed between 2007 and 2016. The per-patient CTC positive rate was 9.8% for polyps from 6 to 9 mm, 5.4% for polyps from 10 to 29 mm, and 1.3% for masses ≥ 30 mm. Based on CTC findings and patient preference, 13.9% of patients were referred to optical colonoscopy and 2.9% were referred to CTC follow-up. Of patients recommended to undergo colonoscopy, 80.6% completed the procedure and colonic pathology was confirmed in 83.2% thereof. Histologic diagnoses were 221 tubular adenomas (161 6-9 mm; 60 ≥ 10 mm), 59 tubulovillous adenomas/villous adenomas (14 6-9 mm; 45 ≥ 10 mm), 34 hyperplastic-serrated polyps (27 6-9 mm; 7 ≥ 10 mm), 18 adenocarcinomas (all ≥ 10 mm), and 11 adenomas with high-grade dysplasia (all ≥ 10 mm). The per-patient positivity rate for patients who had optical colonoscopy (n = 279) was 71.3% for adenoma, 9.7% for hyperplastic polyp, and 6.1% for adenocarcinoma. CONCLUSIONS: CTC is an effective screening modality for African-American adults and could potentially improve rates of colorectal screening in this underserved population.

Automated High-Throughput Damage Scoring of Zebrafish Lateral Line Hair Cells After Ototoxin Exposure.

Zebrafish have emerged as a powerful biological system for drug development against hearing loss. Zebrafish hair cells, contained within neuromasts along the lateral line, can be damaged with exposure to ototoxins, and therefore, pre-exposure to potentially otoprotective compounds can be a means of identifying promising new drug candidates. Unfortunately, anatomical assays of hair cell damage are typically low-throughput and labor intensive, requiring trained experts to manually score hair cell damage in fluorescence or confocal images. To enhance throughput and consistency, our group has developed an automated damage-scoring algorithm based on machine-learning techniques that produce accurate damage scores, eliminate potential operator bias, provide more fidelity in determining damage scores that are between two levels, and deliver consistent results in a fraction of the time required for manual analysis. The system has been validated against trained experts using linear regression, hypothesis testing, and the Pearson's correlation coefficient. Furthermore, performance has been quantified by measuring mean absolute error for each image and the time taken to automatically compute damage scores. Coupling automated analysis of zebrafish hair cell damage to behavioral assays for ototoxicity produces a novel drug discovery platform for rapid translation of candidate drugs into preclinical mammalian models of hearing loss.

Structured Protocol for Benign Biliary Anastomotic Strictures: Impact on Long-Term Clinical Effectiveness.

OBJECTIVE: The purpose of this study is to compare long-term clinical effectiveness before and after implementation of a structured protocol for percutaneous drainage of benign anastomotic biliary strictures. MATERIALS AND METHODS: Three hundred five adult patients undergoing percutaneous biliary drainage for biliary anastomotic strictures between 1994 and 2015 were identified using Current Procedural Terminology billing codes, with 234 undergoing intervention before implementation of a structured protocol and 71 undergoing intervention after implementation of the protocol. The frequency of surgical anastomotic revision was compared between patients treated before and after the implementation of the structured protocol. Patient characteristics and treatment variables were also analyzed with respect to the frequency of surgical revision. A Kaplan-Meier analysis was performed to determine the long-term probabilities of avoiding surgical revision and patency rates. RESULTS: Overall, 72.8% of patients avoided surgical revision, with 71.1% before and 81.7% after the protocol was implemented (p = 0.1052). A larger maximum drain size was significantly associated with a lower frequency of surgical revision (p = 0.0006). The rates of surgical avoidance 5 years after treatment before and after protocol implementation were 69.1% and 80.8%, respectively. Patency rates 5 years after treatment before and after protocol implementation were 73.8% and 76.8%, respectively. CONCLUSION: Percutaneous drainage and management of benign biliary anastomotic strictures is an effective treatment regardless of the presence of a structured protocol. Although there was no significant benefit in terms of avoidance of surgical revision, the time until surgical revision and patency rates were increased with the protocol. In addition, a larger maximum drain size was associated with a better outcome.

"Partially Matched" US Senior Diagnostic Radiology Applicants: Scope of the Problem and Implications for Applicants, Residency Training Programs, and the Academic Diagnostic Radiology Community.

RATIONALE AND OBJECTIVES: Anecdotal reports have recently surfaced of an increase in the number of US senior diagnostic radiology (DR)-bound residency applicants who did not secure a preliminary year position through the Main Resident Match (the Match) of the National Resident Matching Program (NRMP), so-called "partial matches." This study sought to determine the scope of this phenomenon and address potential causes and solutions. MATERIALS AND METHODS: Publicly available and proprietary data from the National Resident Matching Program were analyzed from 2005-2016 to determine the number of partially matched US senior applicants, selectivity of DR residency training, availability of preliminary year positions, number of unique preliminary year applicants, distribution of preliminary year matches by successfully matched applicants in relevant specialties (DR, anesthesiology, dermatology, neurology, physical medicine and rehabilitation, and radiation oncology), and percentage of categorical training programs participating in the Match. RESULTS: Since 2012, there has been a trend toward an increase in the number of partially matched US senior DR applicants, with a transitory recovery in the 2015 Match. Although possible explanations for this increase are proposed, a definitive etiology remains elusive. Strategies to offset this phenomenon include advising perceived at-risk applicants to apply more broadly to preliminary year positions, creating more categorical training positions, abolishing the preliminary year requirement, and drawing high-caliber medical students to the field. CONCLUSIONS: Although a definitive explanation for a recent increase in partial matches remains elusive, strategies exist for DR applicants, residency programs, and the DR academic community to minimize this risk.

Right ventricular functional analysis utilizing first pass radionuclide angiography for pre-operative ventricular assist device planning: a multi-modality comparison.

BACKGROUND: Advanced heart failure treated with a left ventricular assist device is associated with a higher risk of right heart failure. Many advanced heart failures patients are treated with an ICD, a relative contraindication to MRI, prior to assist device placement. Given this limitation, left and right ventricular function for patients with an ICD is calculated using radionuclide angiography utilizing planar multigated acquisition (MUGA) and first pass radionuclide angiography (FPRNA), respectively. Given the availability of MRI protocols that can accommodate patients with ICDs, we have correlated the findings of ventricular functional analysis using radionuclide angiography to cardiac MRI, the reference standard for ventricle function calculation, to directly correlate calculated ejection fractions between these modalities, and to also assess agreement between available echocardiographic and hemodynamic parameters of right ventricular function. METHODS: A retrospective review from January 2012 through May 2014 was performed to identify advanced heart failure patients who underwent both cardiac MRI and radionuclide angiography for ventricular functional analysis. Nine heart failure patients (8 men, 1 woman; mean age of 57.0 years) were identified. The average time between the cardiac MRI and radionuclide angiography exams was 38.9 days (range: 1 - 119 days). All patients undergoing cardiac MRI were scanned using an institutionally approved protocol for ICD with no device-related complications identified. A retrospective chart review of each patient for cardiomyopathy diagnosis, clinical follow-up, and echocardiogram and right heart catheterization performed during evaluation was also performed. RESULTS: The 9 patients demonstrated a mean left ventricular ejection fraction (LVEF) using cardiac MRI of 20.7% (12 - 40%). Mean LVEF using MUGA was 22.6% (12 - 49%). The mean right ventricular ejection fraction (RVEF) utilizing cardiac MRI was 28.3% (16 - 43%), and the mean RVEF calculated by FPRNA was 32.6% (9 - 56%). The mean discrepancy for LVEF between cardiac MRI and MUGA was 4.1% (0 - 9%), and correlation of calculated LVEF using cardiac MRI and MUGA demonstrated an R of 0.9. The mean discrepancy for RVEF between cardiac MRI and FPRNA was 12.0% (range: 2 - 24%) with a moderate correlation (R = 0.5). The increased discrepancies for RV analysis were statistically significant using an unpaired t-test (t = 3.19, p = 0.0061). Echocardiogram parameters of RV function, including TAPSE and FAC, were for available for all 9 patients and agreement with cardiac MRI demonstrated a kappa statistic for TAPSE of 0.39 (95% CI of 0.06 - 0.72) and for FAC of 0.64 (95% of 0.21 - 1.00). CONCLUSION: Heart failure patients are increasingly requiring left ventricular assist device placement; however, definitive evaluation of biventricular function is required due to the increased mortality rate associated with right heart failure after assist device placement. Our results suggest that FPRNA only has a moderate correlation with reference standard RVEFs calculated using cardiac MRI, which was similar to calculated agreements between cardiac MRI and echocardiographic parameters of right ventricular function. Given the need for identification of patients at risk for right heart failure, further studies are warranted to determine a more accurate estimate of RVEF for heart failure patients during pre-operative ventricular assist device planning.

Ultrasound Evaluation of Morton Neuroma Before and After Laser Therapy.

OBJECTIVE: The objective of our study was to retrospectively assess for differences in imaging appearances of Morton neuromas before and after laser therapy using diagnostic ultrasound (US). MATERIALS AND METHODS: A retrospective review was performed to identify patients who underwent US imaging to evaluate for Morton neuroma during the study period (June 1, 2013-July 1, 2014); of the 42 patients identified, 21 underwent US evaluations before and after laser therapy. US reports and images were reviewed and correlated with clinical history. The final study group consisted of 21 patients who had a total of 31 Morton neuromas evaluated using US after treatment. A retrospective review was then performed to characterize the appearances of these lesions before and after therapy followed by an analysis of variables. RESULTS: Retrospective US review of 31 pretreatment Morton neuromas showed fusiform, heterogeneously hypoechoic masses with well-defined borders in most cases and that pain was reported when transducer pressure was applied in 97% (30/31) of cases. After treatment, lesions showed ill-defined borders (23/31), and pain with application of transducer pressure was either significantly decreased or absent (29/31); these findings were concordant with the clinical findings. Both of these characteristics were statistically significant (p < 0.0001). In addition, more Morton neuromas occurred in the second intermetatarsal space than in the third intermetatarsal space (p < 0.0001). CONCLUSION: US may be used to identify posttreatment changes after laser therapy of Morton neuromas. Posttreatment changes include ill-defined borders and less pain or the absence of pain with the application of transducer pressure. These criteria may be applied in future clinical studies evaluating the efficacy of laser therapy for Morton neuroma.