About

James Carr is the Drs. Frederick John Bradd and William Kennedy Memorial Professor of Radiology and a Professor of Biomedical Engineering at Northwestern University. His educational background includes an MD from the Royal College of Surgeons, Ireland, obtained in 1992, followed by residency at Saint Vincent's Hospital, Ireland, completed in 1999, and a fellowship at Northwestern McGaw/Northwestern Memorial Hospital in 2001. His research interests focus on cardiac magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), coronary CT angiography, cardiac CT, CT angiography, and vascular and interventional radiology. Carr has contributed to the development of methodologies for detecting abnormal relative wall shear stress on the thoracic aorta using four-dimensional flow MRI and has explored the influence of aortic valve morphology on aortic hemodynamics. He has authored chapters on MRI of aortic disease and has been involved in advancing imaging techniques for cardiovascular applications.

Research topics

• Medicine
• Cardiology
• Computer Science
• Artificial Intelligence
• Radiology
• Nuclear medicine
• Internal medicine
• Pathology
• Intensive care medicine
• Medical emergency
• Family medicine
• Mathematics
• Urology
• Computer vision
• Medical physics

Selected publications

• Generative AI and Foundation Models in Radiology: Applications, Opportunities, and Potential Challenges

  Radiology · 2025-11-01 · 5 citations

  article

  Foundation models (FMs) represent a transformative advancement in artificial intelligence (AI), with growing applications in medical imaging. These models leverage self-attention mechanisms and are capable of processing multimodal data, such as images, text, audio, and video, across multiple scales. Although FMs require large datasets for initial training, they can be adapted to specific medical imaging tasks using smaller labeled datasets through techniques such as transfer learning, fine-tuning, prompt engineering, few-shot learning, and zero-shot learning, making them especially valuable in data-scarce settings. Many FMs also incorporate generative AI capabilities that support the creation of synthetic medical images to further address annotation limitations. Current applications span various imaging modalities in radiology, where FMs have shown potential to improve diagnostic accuracy and streamline workflows. However, clinical integration remains challenging due to issues such as limited interpretability, potential bias, privacy concerns, regulatory constraints, high computational costs, and domain shifts between training data and real-world clinical environments. Addressing these barriers will require coordinated efforts among technical developers, health care providers, and regulatory bodies. This review explores the evolving role of FMs and generative AI in radiology, highlighting recent research advances, clinical applications, and the key challenges that must be addressed for responsible deployment.

  PublisherDOI

• Cine MRI-Derived Radiomics for the Detection of Functional Tricuspid Regurgitation in Pulmonary Hypertension: A Proof-of-Concept Study

  Journal of Cardiovascular Development and Disease · 2025-09-13

  articleOpen accessSenior author

  (1) Objective: The objective was to test the hypothesis that cine MRI-derived radiomic features can detect functional tricuspid regurgitation (FTR) in the context of pulmonary hypertension (PH). (2) Materials and methods: In total, 53 PH patients were retrospectively enrolled. Thirty-three patients had echocardiography-defined mild-to-severe FTR, while the other twenty patients had no or trivial regurgitation. For all participants, 93 radiomic features were extracted from four-chamber cine MRI using a fixed-size region of interest (ROI) located in the right atrium (RA), 0.5–1 cm above the tricuspid valve. The levels of radiomic features were averaged over the ventricular systole and compared between patients with and without FTR using t tests. In patients with FTR, radiomic features were related to hemodynamic parameters in the right heart using the Pearson correlation coefficient (r). (3) Results: There were no significant differences in demographic information, right heart catheterization (RHC) results, and most cine MRI-derived cardiac function indices between the two subject groups. Eight of ninety-three radiomic features were significantly different between PH patients with and without FTR. Radiomic features can be used to discriminate two subject groups (AUC = 0.77). In patients with FTR, multiple radiomic features are related to the pressure in the RA, right ventricle (RV), and pressure difference between RA and RV (r: 0.4 to 0.55), p values < 0.05. (4) Conclusion: Cine MRI-derived radiomic features of the cardiac blood pool differ between PH patients with and without FTR. Cine MRI shows promise as a method for assessing FTR in the context of complex cardiovascular diseases (CVDs).

  PublisherOA PDFDOI

• Abstract 4369636: Multi-parametric Cardiac MRI predicts cardiovascular outcomes in HTx recipients after long-term follow-up

  Circulation · 2025-11-03

  article

  Background: The long-term survival of heart transplantation (HTx) recipients is influenced by a range of cardiovascular, immunological, and procedural factors. Accurately predicting post-HTx outcomes remains a major clinical challenge, especially when relying solely on noninvasive methods. Objective: To test the hypothesis that structural and functional indices derived from multi-parametric cardiac MRI-derived can be used to predict cardiovascular events in HTx recipients. Materials and methods: With the approval of institutional review board (IRB), 170 HTx recipients (106 males, age: 47.8 ± 16 years, Range: 19 – 79 years) were recruited for a comprehensive multi-parametric cardiac MRI scan. MRI images were processed to derive global cardiac function and volumes, and myocardial T2 values and T1 values. Pre- and post-Gadolinium T1 was used to calculate extra-cellular volume (ECV) fraction. Cardiovascular events were defined as a composite of any emergency visit, hospitalization or death due to graft failure or reception, myocardial infarction, HF and other events that cannot rule out a cardiovascular origin of complications. Identification of predictors of adverse outcomes at long-term follow-up was based on a Cox proportional hazards model (CPH). Statistical analysis was performed by using SPSS (version 22.0). Results: MRI images were eligilbe for quantitative analysis. See figure 1. The patients were followed for 6 to 4504 days (Median = 2616 days) after multi-parametric cardiac MRI. In total, 140 cardiovascular events occurred (6 to 3294 days, Median = 627 days). The CPH model fits the data (p < 0.001). After the adjustment of traditional cardiovascular risk factors and demographic data, multiple MRI-derived indices were identified as significant predictors of survival time (time between baseline cardiac MRI and adverse event), including left ventricular (LV) end-diastolic volume (LVEDV) (p < 0.001), LV end-systolic volume (LVESV) (p < 0.001), LV stroke volume (LVSV) (p = 0.005), right ventricular (RV) stroke volume (RVSV) (p < 0.001), RV cardiac output (RVCO) ( p = 0.03), myocardial ECV (p < 0.001) and T2 value (p = 0.008). See figure 2. Conclusions: Multi-parametric indices of cardiac tissue (T2, ECV) and function (LVEDV, LVESV, LVSV, RVSV, RVCO) can independently predict adverse clinical outcomes in HTx recipients at long term follow-up (median > 7 years). MRI may offer new imaging biomarkers for early identification of risks for post-HTx complication.

  PublisherDOI

• Findings from 4D-flow MRI in an adult case with scimitar sign

  Journal of Cardiovascular Magnetic Resonance · 2025-01-01

  articleOpen accessSenior author
  PublisherDOI

• State of the Art in Pulmonary Arterial Hypertension: Molecular Basis, Imaging Modalities, and Right Heart Failure Treatment

  Biomedicines · 2025-07-20 · 5 citations

  reviewOpen accessSenior authorCorresponding

  Pulmonary hypertension (PH) is broadly defined as a mean pulmonary arterial pressure (mPAP) exceeding 20 mm Hg at rest. Pulmonary arterial hypertension (PAH) is a specific subset of PH characterized by a normal pulmonary arterial wedge pressure (PAWP), combined with elevated mPAP and increased pulmonary vascular resistance (PVR), without other causes of pre-capillary hypertension such as lung diseases or chronic thromboembolic pulmonary hypertension. The majority of PAH cases are idiopathic; other common etiologies include connective tissue disease-associated PAH, congenital heart disease, and portopulmonary hypertension. To a lesser extent, genetic and familial forms of PAH can also occur. The pathophysiology of PAH involves the following four primary pathways: nitric oxide, endothelin-1, prostacyclin, and activin/bone morphogenetic protein (BMP). Dysregulation of these pathways leads to a progressive vasculopathy marked by vasoconstriction, vascular proliferation, elevated right heart afterload, and ultimately right-sided heart failure. Diagnosing PAH is challenging and often occurs at advanced stages. The gold standard for diagnosis remains invasive right heart catheterization. Along with invasive hemodynamic measurements, several noninvasive imaging modalities such as echocardiography and ventilation-perfusion scanning are key adjunct techniques. Also, recent advancements in cardiac magnetic resonance (CMR) have opened a new era for PAH management. Additionally, CMR and echocardiography not only enable diagnosis but also aid in evaluating disease severity and monitoring treatment responses. Current PAH treatments focus on targeting molecular pathways, reducing inflammation, and inhibiting right-sided heart failure. Integrating imaging with basic science techniques is crucial for enhanced patient diagnosis, and precision medicine is emerging as a key strategy in PAH management. Additionally, the incorporation of artificial intelligence into both molecular and imaging approaches holds significant potential. There is a growing need to integrate new imaging modalities with high resolution and reduced radiation exposure into clinical practice. In this review, we discuss the molecular pathways involved in PAH, the imaging modalities utilized for diagnosis and monitoring, and current targeted therapies. Advances in molecular understanding and imaging technologies, coupled with precision medicine, could hold promise in improving patient outcomes and revolutionizing the management of PAH patients.

  PublisherOA PDFDOI

• Multiparametric quantitative structural and functional cardiac MRI in orthotopic heart transplant recipients with cardiac allograft vasculopathy

  The International Journal of Cardiovascular Imaging · 2025-04-17

  articleOpen access

  The aim of this study was to verify if multiparametric quantitative CMR can detect mild-to-moderate cardiac allograft vasculopathy (CAV) in patients post-orthotopic heart transplant (OHT). 51 patients (age = 50.0 ± 13.6 years, 29% female) post-OHT 0-6 years (mean 3.2 ± 1.5 years) who underwent CMR from 2011 to 2019 were retrospectively included. Multiparametric CMR included CINE imaging covering the left ventricle (LV), pre- and post-contrast T1 mapping, and T2 mapping, extracellular volume fraction (ECV) calculation, and 2D-feature tracking strain. CAV0 ('CAV negative') patient variables were compared with CAV1-CAV2 ('CAV positive') variables. Logistic regression was used to determine predictors of CAV status. Myocardial T2 was higher in CAV positive compared with CAV negative patients (54.5 ± 7.7 ms vs. 50.2 ± 3.3 ms, p < 0.05), as was ECV (31.3 ± 5.3% vs. 27.4 ± 4.1%, p < 0.05). Radial and circumferential peak systolic strain rates were attenuated in CAV positive vs. CAV negative patients (radial: 1.4 ± 0.4 s-1 vs. 1.8 ± 0.3 s-1, circumferential: -0.9 ± 0.2 s-1 vs. -1.1 ± 0.1 s-1, p < 0.05),as well as circumferential and longitudinal peak diastolic strain rates (0.7 ± 0.7 s-1 vs. 1.0 ± 0.5 s-1, and 0.8 ± 0.3 s-1 vs. 0.9 ± 0.3 s-1, p < 0.05, respectively). CAV positive vs. negative status correlated with ECV (rho 0.41, P < 0.01), T2 (rho 0.29, p < 0.05), radial and circumferential peak systolic strain rate (rho - 0.48, P < 0.01 and rho 0.47, p < 0.001, respectively), and circumferential and longitudinal peak diastolic strain rates (rho - 0.34, p < 0.05 and rho - 0.35, p < 0.01, respectively). Logistic regression revealed that a model including ECV, peak radial and circumferential systolic strain rates and longitudinal diastolic strain rate was significant for distinguishing CAV positive vs. negative status with a receiver operator characteristic area under curve of 0.85 ± 0.06 (CI 0.73-0.97), p < 0.005. A model combining functional (strain) and tissue parameters (ECV) was predictive of CAV status, indicating the potential of multiparametric CMR for non-invasive prediction of CAV status in OHT recipients.

  PublisherOA PDFDOI

• Inter-AI Agreement in Measuring Cine MRI-Derived Cardiac Function and Motion Patterns: A Pilot Study

  Journal of Imaging Informatics in Medicine · 2025-07-08

  articleSenior author
  PublisherDOI

• Cardiac MRI in Heart Transplantation: Approaches and Clinical Insights

  Radiographics · 2025-01-30 · 6 citations

  review

  Cardiac MRI provides comprehensive insights into orthotopic heart transplant (OHT) allograft structure and function with advanced tissue characterization and quantitative myocardial perfusion techniques and can be considered a complementary noninvasive tool for investigation and monitoring of major complications that limit the durability of OHT.

  PublisherDOI

• Closing the Loop: 9-Year Outcomes of an Electronic Medical Record–Based Protocol for Reporting Incidental Imaging Findings

  Journal of the American College of Radiology · 2025-09-04

  articleSenior author
  PublisherDOI

• Exploiting Machine Learning to Develop Ocean Color Retrievals from the Tropospheric Emissions: Monitoring of Pollution Instrument

  2025-03-20

  preprintOpen access

  • The TEMPO instrument is an ideal proxy to explore future capabilities of planned geostationary ocean color instruments GLIMR and GeoXO OCX • Machine learning can be utilized to develop efficient retrievals for value-added ocean color products

  PublisherDOI

Recent grants

• NIH Grant R01HL089695

  NIH · $1.8M · 2014

• Comprehensive Cardiac Structure-Function Analysis in Heart Transplantation

  NIH · $6.4M · 2014–2027

Frequent coauthors

• Tamás Varga

  Environmental Molecular Sciences Laboratory

  381 shared

• Montana Smith

  Environmental Molecular Sciences Laboratory

  381 shared

• Rey Huachambé

  Pacific Northwest National Laboratory

  381 shared

• Nikola Tolić

  Pacific Northwest National Laboratory

  381 shared

• Thomas Wietsma

  Pacific Northwest National Laboratory

  381 shared

• Arunima Bhattacharjee

  Environmental Molecular Sciences Laboratory

  381 shared

• Alexis Heath

  Pacific Northwest National Laboratory

  381 shared

• Ruonan Wu

  Pacific Northwest National Laboratory

  381 shared

Awards & honors

• Drs. Frederick John Bradd and William Kennedy Memorial Profe…
• Northwestern McGaw/Northwestern Memorial Hospital Fellowship…