Faculty Bibliography

RATIONALE AND OBJECTIVES: This study aimed to characterize recent National Institutes of Health (NIH) funding for diagnostic radiology departments at US medical schools. MATERIALS AND METHODS: This retrospective study did not use private identifiable information and thus did not constitute human subjects research. The public NIH Research Portfolio Online Reporting Tools Expenditure and Results system was used to extract information regarding 887 NIH awards in 2015 to departments of "Radiation-Diagnostic/Oncology." Internet searches were conducted to identify each primary investigator (PI)'s university web page, which was used to identify the PI's departmental affiliation, gender, degree, and academic rank. A total of 649 awards to diagnostic radiology departments, based on these web searches, were included; awards to radiation oncology departments were excluded. Characteristics were summarized descriptively. RESULTS: A total of 61 unique institutions received awards. The top five funded institutions represented 33.6% of all funding. The most common institutes administering these awards were the National Cancer Institute (29.0%) and the National Institute of Biomedical Imaging and Bioengineering (21.6%). Women received 15.9% of awards and 13.3% of funding, with average funding per award of $353,512 compared to $434,572 for men. PhDs received 77.7% of all awards, with average funding per award of $457,413 compared to $505,516 for MDs. Full professors received 51.2% of awards (average funding per award of $532,668), compared to assistant professors who received 18.4% of awards ($260,177). Average funding was $499,859 for multiple-PI awards vs. $397,932 for single-PI awards. Common spending categories included "neurosciences," "cancer," "prevention," and "aging." CONCLUSIONS: NIH funding for diagnostic radiology departments has largely been awarded to senior-ranking male PhD investigators, commonly at large major academic medical centers. Initiatives are warranted to address such disparities and promote greater diversity in NIH funding among diagnostic radiology investigators.

RATIONALE AND OBJECTIVES: This study aims to assess the impact of off-campus facility expansion by a large academic health system on patient travel times for screening mammography. MATERIALS AND METHODS: Screening mammograms performed from 2013 to 2015 and associated patient demographics were identified using the NYU Langone Medical Center Enterprise Data Warehouse. During this time, the system's number of mammography facilities increased from 6 to 19, reflecting expansion beyond Manhattan throughout the New York metropolitan region. Geocoding software was used to estimate driving times from patients' homes to imaging facilities. RESULTS: For 147,566 screening mammograms, the mean estimated patient travel time was 19.9 +/- 15.2 minutes. With facility expansion, travel times declined significantly (P < 0.001) from 26.8 +/- 18.9 to 18.5 +/- 13.3 minutes (non-Manhattan residents: from 31.4 +/- 20.3 to 18.7 +/- 13.6). This decline occurred consistently across subgroups of patient age, race, ethnicity, payer status, and rurality, leading to decreased variation in travel times between such subgroups. However, travel times to pre-expansion facilities remained stable (initial: 26.8 +/- 18.9 minutes, final: 26.7 +/- 18.6 minutes). Among women undergoing mammography before and after expansion, travel times were shorter for the postexpansion mammogram in only 6.3%, but this rate varied significantly (all P < 0.05) by certain demographic factors (higher in younger and non-Hispanic patients) and was as high as 18.2%-18.9% of patients residing in regions with the most active expansion. CONCLUSIONS: Health system mammography facility geographic expansion can improve average patient travel burden and reduce travel time variation among sociodemographic populations. Nonetheless, existing patients strongly tend to return to established facilities despite potentially shorter travel time locations, suggesting strong site loyalty. Variation in travel times likely relates to various factors other than facility proximity.

PURPOSE: To use patient-generated data to assess the changing role of emergency department (ED) imaging for a spectrum of clinical indications. METHODS: The Household Component Emergency Room Visits File was obtained from 1996 through 2014 for the Medical Expenditure Panel Survey, a nationally representative survey of US households. Percentage of visits associated with various imaging modalities was computed annually, stratified by respondents' self-reported primary condition during the visit. Modality characteristics were assessed for conditions most frequently imaged in 1996 or 2014. RESULTS: For most conditions, use of advanced imaging (defined by Medical Expenditure Panel Survey as CT or MRI) in the ED increased significantly (P < .001). The largest growth occurred for urinary calculus (from 0% to 48.5%) and headache (from 17.5% to 33.3%), which were the most commonly imaged conditions by CT or MRI in 2014. For ultrasound, the most commonly imaged condition was pregnancy in 1996 (32.9%) and 2014 (44.5%). No other condition was associated with ultrasound in >20% of visits. For radiography, the most commonly imaged conditions were extremity wounds and fractures in 1996 (range 84.5%-90.2%) and 2014 (range 93.4%-93.9%). Use of radiography decreased for urinary calculus from 67.4% to 24.2% (P < .001). CONCLUSION: For many conditions, ED utilization of advanced imaging increased significantly, though growth was variable across conditions. In certain scenarios, advanced ED imaging is adding to, rather than replacing, other modalities. Ultrasound and radiography utilization was overall unchanged. That national patient survey data mirror traditional claims-based studies suggests an expanded role for patient-generated data in identifying areas of imaging utilization that may benefit from targeted optimization efforts.

Purpose To compare standard diffusion-weighted (DW) imaging and diffusion kurtosis (DK) imaging for prostate cancer (PC) detection and characterization in a large patient cohort, with attention to the potential added value of DK imaging. Materials and Methods This retrospective institutional review board-approved study received a waiver of informed consent. Two hundred eighty-five patients with PC underwent 3.0-T phased-array coil prostate magnetic resonance (MR) imaging, including a DK imaging sequence (b values 0, 500, 1000, 1500, and 2000 sec/mm2) before prostatectomy. Maps of apparent diffusion coefficient (ADC) and diffusional kurtosis (K) were derived by using maximal b values of 1000 and 2000 sec/mm2, respectively. Mean ADC and K were obtained from volumes of interest (VOIs) placed on each patient's dominant tumor and benign prostate tissue. Metrics were compared between benign and malignant tissue, between Gleason score (GS) /= 3 + 4 tumors, and between GS /= 4 + 3 tumors by using paired t tests, analysis of variance, receiver operating characteristic (ROC) analysis, and exact tests. Results ADC and K showed significant differences for benign versus tumor tissues, GS /= 3 + 4 tumors, and GS /= 4 + 3 tumors (P < .001 for all). ADC and K were highly correlated (r = -0.82; P < .001). Area under the ROC curve was significantly higher (P = .002) for ADC (0.921) than for K (0.902) for benign versus malignant tissue but was similar for GS /= 3 + 4 tumors (0.715-0.744) and GS /= 4 + 3 tumors (0.694-0.720) (P > .15). ADC and K were concordant for these various outcomes in 80.0%-88.6% of patients; among patients with discordant results, ADC showed better performance than K for GS /= 4 + 3 tumors (P = .016) and was similar to K for other outcomes (P > .136). Conclusion ADC and K were highly correlated, had similar diagnostic performance, and were concordant for the various outcomes in the large majority of cases. These observations did not show a clear added value of DK imaging compared with standard DW imaging for clinical PC evaluation. (c) RSNA, 2017 Online supplemental material is available for this article.

The radiologist's role extends far beyond interpretation and reporting of medical imaging. In this manuscript, we describe the role of the Director of Prostate Imaging. We believe that this model can and should be implemented at other institutions, ultimately serving to improve the care for prostate cancer patients. Moreover, this model can be translated to support the development of an array of patient-centered service lines not only in abdominal imaging, but throughout radiology practices at large.

PURPOSE: To explore the role of whole-lesion apparent diffusion coefficient (ADC) analysis for predicting outcomes in prostate cancer patients on active surveillance. METHODS: This study included 72 prostate cancer patients who underwent MRI-ultrasound fusion-targeted biopsy at the initiation of active surveillance, had a visible MRI lesion in the region of tumor on biopsy, and underwent 3T baseline and follow-up MRI examinations separated by at least one year. Thirty of the patients also underwent an additional MRI-ultrasound fusion-targeted biopsy after the follow-up MRI. Whole-lesion ADC metrics and lesion volumes were computed from 3D whole-lesion volumes-of-interest placed on lesions on the baseline and follow-up ADC maps. The percent change in lesion volume on the ADC map between the serial examinations was computed. Statistical analysis included unpaired t tests, ROC analysis, and Fisher's exact test. RESULTS: Baseline mean ADC, ADC0-10th-percentile, ADC10-25th-percentile, and ADC25-50th-percentile were all significantly lower in lesions exhibiting >/=50% growth on the ADC map compared with remaining lesions (all P /=50% growth observed for ADC0-10th-percentile (585 +/- 308 vs. 911 +/- 336; P = 0.001). ADC0-10th-percentile achieved highest performance for predicting >/=50% growth (AUC = 0.754). Mean percent change in tumor volume on the ADC map was 62.3% +/- 26.9% in patients with GS >/= 3 + 4 on follow-up biopsy compared with 3.6% +/- 64.6% in remaining patients (P = 0.050). CONCLUSION: Our preliminary results suggest a role for 3D whole-lesion ADC analysis in prostate cancer active surveillance.

PURPOSE: To examine changes in the utilization of procedures related to treatment of chronic venous insufficiency (CVI) in the Medicare population. MATERIALS AND METHODS: Service-specific claims data for phlebectomy, sclerotherapy, and radiofrequency (RF) and laser ablation were identified by using Medicare Physician Supplier Procedure Summary master files from 2005 through 2014. Longitudinal national utilization rates were calculated by using annual Medicare enrollment data from 2005 through 2013. Procedure volumes by specialty group and site of service were analyzed. RESULTS: Total annual claims for these procedures in the Medicare fee-for-service beneficiaries increased from 95,206 to 332,244 (Compound Annual Growth Rate [CAGR], 15%) between 2005 and 2014. Per 1,000 beneficiaries, overall utilization increased annually from 2.8 in 2005 to 9.4 in 2013. Most procedures were performed in the private office setting (92% in 2014). In 2014, radiologists had a 10% relative market share, compared with vascular surgeons, other surgeons, and cardiologists, who had 26%, 25%, and 14% market shares, respectively. Cardiologists had the fastest relative growth, with a CAGR of 51% compared with 23% for radiology, 12% for vascular surgery, and 13% for other surgery. Total venous RF ablation services grew with a CAGR of 31%, with radiology and cardiology growing most rapidly (40% and 79%, respectively). Total venous laser ablation services grew with a CAGR of 22%, with radiology growing 15% and cardiology growing most rapidly at 44%. CONCLUSIONS: Utilization of CVI procedures in the Medicare population increased markedly from 2005 through 2014. The overwhelming majority are performed in the private office setting by nonradiologists.