Mark A Fogel

The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, United States

Are you Mark A Fogel?

Claim your profile

Publications (174)786.83 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: -Patients with repaired tetralogy of Fallot (TOF) experience variable outcomes for reasons that are incompletely understood. We hypothesize that genetic variants contribute to this variability. We sought to investigate the association of 22q11.2 deletion status with clinical outcome in patients with repaired TOF. -We performed a cross sectional study of TOF subjects who were tested for 22q11.2 deletion, and underwent cardiac magnetic resonance (CMR), exercise stress test (EST) and review of medical history. We studied 165 subjects (12.3 ± 3.1 years), of which 30 (18%) had 22q11.2 deletion syndrome (22q11.2DS). Overall, by CMR the right ventricular (RV) ejection fraction was 60±8%, pulmonary regurgitant fraction 34±17%, and RV end-diastolic volume 114±39 cc/m(2). On EST, maximum oxygen consumption (mVO2) was 76±16% predicted. Despite comparable RV function and pulmonary regurgitant fraction, on EST the 22q11.2DS had significantly lower percent predicted: forced vital capacity (61.5 ± 16 vs. 80.5 ± 14, p< 0.0001); mVO2 (61±17 vs. 80±12, p<0.0001); and work (64±18 vs. 86±22, p=0.0002). Similarly, the 22q11.2DS experienced more hospitalizations (6.5 [5; 10] vs. 3 [2; 5], p<0.0001), saw more specialists (3.5 [2; 9] vs. 0 [0; 12], p<0.0001) and used one or more medications (67 vs. 34%, p <0.001). -22q11.2DS is associated with restrictive lung disease, worse aerobic capacity, and increased morbidity, and may explain some of the clinical variability seen in TOF. These findings may provide avenues for intervention to improve outcomes, and should be re-evaluated longitudinally as these associations may become more pronounced with time.
    Circulation Cardiovascular Genetics 01/2015; · 5.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE. Anomalous left coronary artery from the inappropriate aortic sinus with intraseptal course is generally benign but can be confused on imaging studies with the potentially lethal interarterial, intramural anomalous left coronary artery. The purpose of this study was to assess normal ostial morphologic features and intraseptal course using cardiac MRI and CT in pediatric patients with intraseptal anomalous left coronary artery. MATERIALS AND METHODS. A retrospective review was conducted of the medical records of 14 children with the diagnosis of intraseptal anomalous left coronary artery between November 2009 and March 2013. Coronary artery origin and course were evaluated with cardiac MRI or CT, and 3D assessment of coronary ostial morphologic features was performed with virtual angioscopy. RESULTS. The patient ages ranged from 5 to 18 years at diagnosis; 10 (71.4%) were boys. The right and left coronary origins were the right sinus of Valsalva as a common origin (n = 9) or a single coronary artery (n = 5). Anomalous intraseptal left main coronary was found in 13 patients, and one patient had anomalous left anterior descending with retroaortic circumflex coronary artery. Anomalous coronary ostia were round and without stenosis in all studies. The anomalous vessel was identified with echocardiography, but the anomalous left coronary artery was not delineated, and a normal ostium was not adequately portrayed in any instance. CONCLUSION. By use of cardiac MRI and CT, the anomalous course of round coronary ostia was confirmed and visualized in a pediatric cohort with intraseptal anomalous left coronary artery. The data provide the basis for understanding the benign clinical course and showing that surgery is unnecessary for this coronary anomaly.
    American Journal of Roentgenology 01/2015; 204(1):W104-9. · 2.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Utilization of cardiovascular magnetic resonance (CMR) is limited in young children because of the need for sedation or general anesthesia (GA). It has been previously shown that CMR can be performed without sedation or GA in young infants who are prone to fall asleep after being fed and swaddled. The purpose of this study was to prospectively prove the feasibility of the feed-and-sleep CMR technique in larger cohorts in the two institutions where the technique was initially developed. This was a prospective dual-center cohort study over a two-year period. All infants younger than 6 months old with complex congenital cardiovascular anomaly who required CMR were recruited for this study. The exclusion criteria included mechanical ventilation, oxygen dependence, feeding difficulties, and any contraindication to CMR. The feed-and-sleep study was performed by fasting the infant for a period of 4 h prior to the scan, placing the infant in a vacuum immobilizer, and feeding the infant just prior to the CMR. The CMR sequences were prioritized to target the area of most importance first. A study was considered complete and diagnostic if the clinical question was answered. A total of 60 infants (39 from center A and 21 from center B) were recruited for this study, 32 male and 28 female, ages ranging from 1 to 177 days (50 ± 54). The CMR studies were diagnostic and answered the clinical questions in all patients. All infants tolerated the procedure well, and no complications were noted in any of the patients. The CMR duration ranged between 4-132 minutes (45 ± 21). The feed-and-sleep approach in selected patients obviates the need of sedation or GA for CMR in infants younger than 6 months old. Therefore, CMR can be utilized whenever echocardiography fails to provide the complete information required for the patients' management.
    Pediatric Cardiology 12/2014; · 1.55 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Pulmonary insufficiency (PI) is associated with right ventricular (RV) dilation, dysfunction, and exercise intolerance in patients with tetralogy of fallot (TOF). We sought to compare RV function and exercise performance in patients with valvar pulmonary stenosis (VPS) following pulmonary balloon valvuloplasty to those with repaired TOF with similar degrees of PI. We performed a cross-sectional study of patients with VPS and TOF. Cardiac magnetic resonance (CMR) and exercise stress test were performed. Subjects were matched by time from initial procedure and severity of PI using propensity scores. After matching, there were 16 patients with VPS and 16 with TOF for comparison, with similar demographics. Time from initial procedure was 14 years (12-16), p = 0.92, and pulmonary regurgitant fraction was 19 % (6-31), p = 0.94, Patients with TOF had lower ejection fraction [58 % (53-66) vs. 65 % (60-69), p = 0.04] and more RV hypertrophy [69 g/m(2) (52-86) vs. 44 g/m(2) (32-66), p = 0.04] compared to those with VPS. Aerobic capacity was worse in patients with TOF [68 ± 19 % mVO2 (56-84) vs. 82 ± 9.2 % (74-89) in VPS, p = 0.01], with a trend for less habitual physical activity [0.9 (0-12) vs. 8 h/week (4-12), p = 0.056], respectively. With similar degrees of PI, patients with TOF demonstrate worse RV function and aerobic capacity as compared to patients with just VPS. Habitual exercise may in part explain differences in exercise performance and should be further explored.
    Pediatric Cardiology 12/2014; · 1.55 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Pulse-wave velocity (PWV), a measure of arterial stiffness, is a known independent risk factor for cardiovascular events. Patients with single ventricle who undergo aortic to pulmonary anastomosis (recon) have noncompliant patch material inserted into the neoaorta, possibly increasing vessel stiffness and afterload. The purpose of this study is to determine if PWV in patients with single ventricle differed between those who did and those who did not undergo aortic reconstruction (nonrecon). We retrospectively reviewed cardiac magnetic resonance anatomic, cine, and phase contrast evaluations in the ascending aorta and descending aorta (DAo) at the level of the diaphragm data from 126 patients with single ventricle (8.6 ± 8.0 years) from January 2012 to May 2013. Significance = p <0.05. Seventy-five patients underwent recon and 51 did not. PWV in recon was significantly higher than in nonrecon (3.9 ± 0.9 m/s vs 3.2 ± 1.0 m/s, p = 0.008); in recon, patients >13 years old had a higher PWV than those <7 years (4.5 ± 0.6 vs 3.5 ± 0.7 m/s, p = 0.004). Whether <7 or >13 years old, PWV of those with recon was higher than nonrecon DAo distensibility was similar between both groups. There was no difference in age, body surface area, or cardiac index between recon and nonrecon. No correlations between various hemodynamic and ventricular function parameters with PWV were noted. In conclusion, PWV in recon is higher than in nonrecon with similar DAo distensibility implicating the aortic reconstruction as a possible cause of increased afterload; older recon patients have stiffer aortas than younger ones, possibly imposing an additional cardiovascular risk in the future. Other biomaterials may potentially moderate PWV if clinical outcome is adversely affected. Copyright © 2014 Elsevier Inc. All rights reserved.
    The American Journal of Cardiology 12/2014; 114(12):1902-7. · 3.43 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: As patients with a single-ventricle physiology age, long-term complications inherent to this population become more evident. Previous studies have focused on correlating anatomic and hemodynamic performance, but there is little information of how these variables change with time. Vessel growth and flow rate changes were quantified using cardiac magnetic resonance and their effects on hemodynamics were assessed, which could affect the long-term outcome. Forty-eight patients with a lateral tunnel or extracardiac conduit Fontan who underwent two cardiac magnetic resonance scans (average interval, 5.1 ± 2.3 years) were studied. Total cavopulmonary connection anatomic and flow variables were reconstructed and normalized to body surface area(1/2). Total cavopulmonary connection hemodynamic efficiency (indexed power loss) was obtained through computational fluid dynamic modeling. Absolute vessel diameters increased with time, normalized diameters decreased, and vessel mean flow rates remained unchanged. Indexed power loss changed significantly in the cohort, as well as in patients in whom the minimum normalized left pulmonary artery decreased. Age at first scan and connection type (lateral tunnel or extracardiac conduit) were not associated with changes in indexed power loss. We present the largest serial cardiac magnetic resonance Fontan cohort to date. Although flow rates increased proportionally to body surface area, vessel diameters did not match somatic growth. As a result, energy losses increased significantly with time in the cohort analyzed. Copyright © 2014 The Society of Thoracic Surgeons. Published by Elsevier Inc. All rights reserved.
    The Annals of Thoracic Surgery 11/2014; · 3.63 Impact Factor
  • 2014 American Academy of Pediatrics National Conference and Exhibition; 10/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Elevated energy loss in the total cavopulmonary connection (TCPC) is hypothesised to have a detrimental effect on clinical outcomes in single-ventricle physiology, which may be magnified with exercise. This study investigates the relationship between TCPC haemodynamic energy dissipation and exercise performance in single-ventricle patients.
    Heart (British Cardiac Society) 09/2014; · 6.02 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Single-ventricle patients undergoing surgical reconstruction experience a high rate of brain injury. Incidental findings on preoperative brain scans may result in safety considerations involving hemorrhage extension during cardiopulmonary bypass that result in surgical postponement.
    The Annals of Thoracic Surgery 08/2014; · 3.63 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The considerable blood mixing in the bidirectional Glenn (BDG) physiology further limits the capacity of the single working ventricle to pump enough oxygenated blood to the circulatory system. This condition is exacerbated under severe conditions such as physical activity or high altitude. In this study, the effect of high altitude exposure on hemodynamics and ventricular function of the BDG physiology is investigated. For this purpose, a mathematical approach based on a lumped parameter model was developed to model the BDG circulation. Catheterization data from 39 BDG patients at stabilized oxygen conditions was used to determine baseline flows and pressures for the model. The effect of high altitude exposure was modeled by increasing the pulmonary vascular resistance (PVR) and heart rate (HR) in increments up to 80% and 40% respectively. The resulting differences in vascular flows, pressures and ventricular function parameters were analyzed. By simultaneously increasing PVR and HR, significant changes (p <0.05) were observed in cardiac index (11% increase at an 80% PVR and 40% HR increase) and pulmonary flow (26% decrease at an 80% PVR and 40% HR increase). Significant increase in mean systemic pressure (9%) was observed at 80% PVR (40% HR) increase. The results show that the poor ventricular function fails to overcome the increased preload and implied low oxygenation in BDG patients at higher altitudes, especially for those with high baseline PVRs. The presented mathematical model provides a framework to estimate the hemodynamic performance of BDG patients at different PVR increments.
    Journal of biomechanics 06/2014; · 2.66 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE. Protein-losing enteropathy (PLE) and plastic bronchitis are serious complications that occur after single-ventricle surgery. A lymphatic cause for these conditions has been proposed, but imaging correlation has not been reported. The objective of this study was to evaluate lymphatic abnormalities in patients after functional single-ventricle palliation compared with patients with non-single-ventricle congenital heart conditions using T2-weighted MR lymphangiography. MATERIALS AND METHODS. We retrospectively reviewed imaging data from 48 patients who underwent T2-weighted MR lymphangiography in our institution between May 1, 2012, and October 24, 2012. The patients were divided into four groups: patients who underwent superior cavopulmonary connection, patients who underwent total cavopulmonary connection, patients with total cavopulmonary connection and lymphatic complications, and patients with non-single-ventricle cardiac anomalies. RESULTS. There were 38 patients with single ventricles in this study. The lymphatic abnormalities observed in these patients included thoracic duct dilation greater than 3 mm (31%), lymphangiectasia and lymphatic collateralization (78%), and tissue edema (86%). There were five patients with PLE, one patient with plastic bronchitis, and one patient with chronic chylous effusions and superior cavopulmonary connection. The patients with PLE and plastic bronchitis had statistically significant larger thoracic duct maximal diameters (median, 3.9 mm; range, 3-7.2 mm) than did the other patients with total cavopulmonary connection (p < 0.01). In the two-ventricle patient group, there were no abnormal lymphatic findings. CONCLUSION. Lymphatic abnormalities are found in many patients after functional single-ventricle palliation. T2-weighted unenhanced MRI is capable of anatomic assessment of the lymphatic system in this patient population and has promise for guiding treatment in the future.
    American Journal of Roentgenology 05/2014; · 2.74 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Tricuspid annular plane systolic excursion (TAPSE) reflects longitudinal myocardial shortening, the main component of right ventricular (RV) contraction in normal hearts. To date, TAPSE has not been extensively studied in patients with hypoplastic left heart syndrome (HLHS) and systemic RVs after Fontan palliation. This retrospective study investigated HLHS patients after Fontan with cardiac magnetic resonance (CMR) performed between 1 January 2010 and 1 August 2012 and transthoracic echocardiogram (TTE) performed within 6 months of CMR. The maximal apical displacement of the lateral tricuspid valve annulus was measured on CMR (using four-chamber cine images) and on TTE (using two-dimensional apical views). To create TTE-TAPSE z-scores, published reference data were used. Intra- and interobserver variability was tested with analysis of variance. Inter-technique agreement of TTE and CMR was tested with Bland-Altman analysis. In this study, 30 CMRs and TTEs from 29 patients were analyzed. The age at CMR was 14.1 ± 7.1 years, performed 11.9 ± 7.8 years after Fontan. For CMR-TAPSE, the intraclass correlation coefficients for inter- and intraobserver variability were 0.89 and 0.91, respectively. The TAPSE measurements were 0.57 ± 0.2 cm on CMR and 0.70 ± 0.2 cm on TTE (TTE-TAPSE z score, -8.7 ± 1.0). The mean difference in TAPSE between CMR and TTE was -0.13 cm [95 % confidence interval (CI) -0.21 to -0.05], with 95 % limits of agreement (-0.55 to 0.29 cm). The study showed no association between CMR-TAPSE and RVEF (R = 0.08; p = 0.67). In patients with HLHS after Fontan, TAPSE is reproducible on CMR and TTE, with good agreement between the two imaging methods. Diminished TAPSE suggests impaired longitudinal shortening in the systemic RV. However, TAPSE is not a surrogate for RVEF in this study population.
    Pediatric Cardiology 05/2014; · 1.55 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Radiation exposure in the pediatric population may increase the risk of future malignancy. Children with congenital heart disease (CHD) who often undergo repeated catheterizations are at risk. One possible strategy to reduce radiation is to use X-ray Magnetic Resonance Fusion (XMRF) to facilitate cardiac catheterization. Methods: Catheterization data of patients who underwent diagnostic XMRF procedures between 1/1/2009 and 2/1/2012 were reviewed. Cases were matched 1:1 to contemporary controls who did not undergo XMRF based on weight and diagnosis and were compared in radiation exposure, contrast dose and procedural and anesthesia times. Results: 44 matched pairs were included. Baseline demographics were similar in both groups. Patients in the XMRF group had lower indices of radiation exposure measured by fluoroscopy time (14 v. 16.4 v. p = 0.047), dose-area product from fluoroscopy (513.2 v. 589.1 µGy·m(2) , p = 0.042), total dose-area product (625.8 v. 995.2 µGy·m(2) , p = 0.027) and total air kerma dose (94.5 v. 153.8 mGy, p = 0.017). There was also a significant reduction in contrast dose (2 v. 3.3 cc/kg, p <0.001). Procedural time tended to be shorter in the XMRF group but anesthesia time was significantly longer. Conclusion: Select diagnostic cardiac catheterization cases that utilized XMRF used less radiation and contrast than similar cases where XMRF was not used. Future work is needed to determine if similar benefits can be extended to other types of diagnostic and complex interventional procedures. © 2014 Wiley Periodicals, Inc.
    Catheterization and Cardiovascular Interventions 03/2014; · 2.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Total cavopulmonary connection (TCPC) geometries have great variability. Geometric features, such as diameter, connection angle, and distance between vessels, are hypothesized to affect the energetics and flow dynamics within the connection. This study aimed to identify important geometric characteristics that can influence TCPC hemodynamics. Anatomies from 108 consecutive patients were reconstructed from cardiac magnetic resonance (CMR) images and analyzed for their geometric features. Vessel flow rates were computed from phase contrast CMR. Computational fluid dynamics simulations were carried out to quantify the indexed power loss and hepatic flow distribution. TCPC indexed power loss correlated inversely with minimum Fontan pathway (FP), left pulmonary artery, and right pulmonary artery diameters. Cardiac index correlated with minimum FP diameter and superior vena cava (SVC) minimum/maximum diameter ratio. Hepatic flow distribution correlated with caval offset, pulmonary flow distribution, and the angle between FP and SVC. These correlations can have important implications for future connection design and patient follow-up.
    JACC. Cardiovascular imaging 03/2014; 7(3). · 14.29 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Typically, a Fontan connection is constructed as either a lateral tunnel (LT) pathway or an extracardiac (EC) conduit. The LT is formed partially by atrial wall and is assumed to have growth potential, but the extent and nature of LT pathway growth have not been well characterized. A quantitative analysis was performed to evaluate this issue. Retrospective serial cardiac magnetic resonance data were obtained for 16 LT and 9 EC patients at 2 time points (mean time between studies, 4.2 ± 1.6 years). Patient-specific anatomies and flows were reconstructed. Geometric parameters of Fontan pathway vessels and the descending aorta were quantified, normalized to body surface area (BSA), and compared between time points and Fontan pathway types. Absolute LT pathway mean diameters increased over time for all but 2 patients; EC pathway size did not change (2.4 ± 2.2 mm vs 0.02 ± 2.1 mm, p < 0.05). Normalized LT and EC diameters decreased, while the size of the descending aorta increased proportionally to BSA. Growth of other cavopulmonary vessels varied. The patterns and extent of LT pathway growth were heterogeneous. Absolute flows for all vessels analyzed, except for the superior vena cava, proportionally to BSA. Fontan pathway vessel diameter changes over time were not proportional to somatic growth but increases in pathway flows were; LT pathway diameter changes were highly variable. These factors may impact Fontan pathway resistance and hemodynamic efficiency. These findings provide further understanding of the different characteristics of LT and EC Fontan connections and set the stage for further investigation.
    The Annals of thoracic surgery 01/2014; · 3.45 Impact Factor
  • Journal of Cardiovascular Magnetic Resonance 01/2014; 16(Suppl 1):P123. · 4.44 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This study sought to quantify average hemodynamic metrics of the Fontan connection as reference for future investigations, compare connection types (intra-atrial vs extracardiac), and identify functional correlates using computational fluid dynamics in a large patient-specific cohort. Fontan hemodynamics, particularly power losses, are hypothesized to vary considerably among patients with a single ventricle and adversely affect systemic hemodynamics and ventricular function if suboptimal. Fontan connection models were created from cardiac magnetic resonance scans for 100 patients. Phase velocity cardiac magnetic resonance in the aorta, vena cavae, and pulmonary arteries was used to prescribe patient-specific time-averaged flow boundary conditions for computational fluid dynamics with a customized, validated solver. Comparison with 4-dimensional cardiac magnetic resonance velocity data from selected patients was used to provide additional verification of simulations. Indexed Fontan power loss, connection resistance, and hepatic flow distribution were quantified and correlated with systemic patient characteristics. Indexed power loss varied by 2 orders of magnitude, whereas, on average, Fontan resistance was 15% to 20% of published values of pulmonary vascular resistance in single ventricles. A significant inverse relationship was observed between indexed power loss and both systemic venous flow and cardiac index. Comparison by connection type showed no differences between intra-atrial and extracardiac connections. Instead, the least efficient connections revealed adverse consequences from localized Fontan pathway stenosis. Fontan power loss varies from patient to patient, and elevated levels are correlated with lower systemic flow and cardiac index. Fontan connection type does not influence hemodynamic efficiency, but an undersized or stenosed Fontan pathway or pulmonary arteries can be highly dissipative.
    The Journal of thoracic and cardiovascular surgery 12/2013; · 3.41 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Fontan procedure, although an imperfect solution for children born with a single functional ventricle, is the only reconstruction at present short of transplantation. The haemodynamics associated with the total cavopulmonary connection, the modern approach to Fontan, are severely altered from the normal biventricular circulation and may contribute to the long-term complications that are frequently noted. Through recent technological advances, spear-headed by advances in medical imaging, it is now possible to virtually model these surgical procedures and evaluate the patient-specific haemodynamics as part of the pre-operative planning process. This is a novel paradigm with the potential to revolutionise the approach to Fontan surgery, help to optimise the haemodynamic results, and improve patient outcomes. This review provides a brief overview of these methods, presents preliminary results of their clinical usage, and offers insights into its potential future directions.
    Cardiology in the Young 12/2013; 23(6):817-822. · 0.95 Impact Factor
  • Circulation Cardiovascular Imaging 11/2013; 6(6):1092-1101. · 5.80 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: -Cardiac catheterization is routinely used as a diagnostic tool in single ventricle patients with superior cavopulmonary connection (SCPC). This physiology presents inherent challenges in applying the Fick principle to estimate flow. We sought to quantitatively define the error in oximetry-derived flow parameters, using phase-contrast cardiac MRI (CMR) as a reference. -Thirty patients with SCPC who underwent combined cardiac MRI and catheterization between July 2008 and June 2012 were retrospectively analyzed. Estimates of flow and resistance calculated using the Fick equation were compared to CMR measurements. Oximetry underestimated CMR-measured pulmonary blood flow (Qp) by an average of 1.1 L/min/m(2), or 32% of the CMR value (p < .0001). Oximetry overestimated systemic blood flow (Qs) by an average of 0.5 L/min/m(2), or 15% of the CMR value (p = .009). There was no correlation between the Qp:Qs ratio derived by Fick and that measured by CMR (ρc = 0.01). The error in Fick Qp correlated moderately with the measured systemic to pulmonary arterial collateral flow (r =0.39). The median total oxygen consumption calculated using combined CMR and oximetry data was 173 mL/min/m(2), higher than the assumed values used to calculate flows by the Fick equation. The upper body circulation received on average 51% of systemic blood flow while conducting only 39% of total body metabolism. -Fick-derived estimates of flow are inherently unreliable in patients with superior cavopulmonary connections. Integrating flows measured by CMR and pressures measured by catheter will provide the best characterization of SCPC physiology.
    Circulation Cardiovascular Imaging 10/2013; · 5.80 Impact Factor

Publication Stats

2k Citations
786.83 Total Impact Points


  • 1995–2014
    • The Children's Hospital of Philadelphia
      • • Division of Cardiology
      • • Department of Radiology
      • • Department of Pediatrics
      • • The Cardiac Center
      Philadelphia, Pennsylvania, United States
  • 2003–2013
    • Georgia Institute of Technology
      • Department of Biomedical Engineering
      Atlanta, GA, United States
  • 2012
    • University of Pennsylvania
      • Department of Physics and Astronomy
      Philadelphia, PA, United States
  • 2011
    • University of Zurich
      Zürich, Zurich, Switzerland
  • 1993–2011
    • Hospital of the University of Pennsylvania
      • • Department of Pediatrics
      • • Department of Radiology
      Philadelphia, Pennsylvania, United States
  • 2009
    • University of Nebraska at Omaha
      Omaha, Nebraska, United States
  • 2008
    • Carnegie Mellon University
      • Department of Biomedical Engineering
      Pittsburgh, PA, United States
  • 2007
    • Yale University
      New Haven, Connecticut, United States