About

Professor Francisco Contijoch is the Principal Investigator of the Contijoch Research Laboratory. The webpage lists him as a key member and leader of the lab, focusing on research related to biomedical engineering and related fields. The lab includes various members such as project scientists, doctoral students, master's students, and support staff, indicating a broad research environment. Specific details about his research focus, background, or key contributions are not provided on the page.

Research topics

• Medicine
• Chemistry
• Biomedical engineering
• Demography
• Internal medicine
• Cell biology
• Biology

Selected publications

• Deep learning enables fully automated cineCT-based assessment of regional right ventricular function

  European Heart Journal - Imaging Methods and Practice · 2026-01-01

  articleOpen accessSenior author

  Aims: Right ventricular (RV) function is a key factor in the diagnosis and prognosis of heart disease. However, current advanced computed tomography (CT)-based assessments rely on semi-automated segmentation of the RV blood pool and manual delineation of the RV free and septal wall boundaries. These steps are time-consuming and prone to inter- and intra-observer variability. Methods and results: We developed and evaluated a fully automated pipeline consisting of two deep learning methods to automate volumetric and regional strain analysis of the RV from contrast-enhanced, electrocardiogram (ECG)-gated cineCT images. The Right Heart Blood Segmenter (RHBS) is a 3D high-resolution configuration of nnU-Net to define the endocardial boundary, while the Right Ventricular Wall Labeler (RVWL) is a 3D point cloud-based deep learning method to label the free and septal walls. We trained our models using a diverse cohort of patients with different RV phenotypes and tested them in an independent cohort of patients with aortic stenosis undergoing TAVR. Our approach demonstrated high accuracy in both cross-validation and independent validation cohorts. RHBS and RVWL both yielded Dice scores of 0.96 and accurate volumetry metrics. RVWL achieved high Dice scores (>0.90) and high accuracy (>93%) for wall labelling. The combination of RHBS + RVWL provided an accurate assessment of free and septal wall regional strain, with a median cosine similarity value of 0.97 in the independent cohort. Conclusion: A fully automated 3D cineCT-based RV regional strain analysis pipeline has the potential to significantly enhance the efficiency and reproducibility of RV function assessment, enabling the evaluation of large cohorts and multi-centre studies.

  PublisherDOI

• In vivo assessment of SNR and v/VENC contributions to velocity uncertainty in PC-MRI

  Journal of Cardiovascular Magnetic Resonance · 2026-01-01

  articleOpen accessSenior author
  PublisherDOI

• Sinogram‑based flow estimation in CT: impact of x‑ray fluence and pulsed acquisition on accuracy and dose

  2026-04-01

  articleSenior author
  PublisherDOI

• Left atrial flow and thrombosis risk from 4D CT contrast dynamics by physics-informed neural network and indicator dilution theory

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-03

  articleOpen access

  Atrial fibrillation (AF) promotes blood stasis and thrombus formation, most often within the left atrial appendage (LAA), and can lead to stroke or transient ischemic attack (TIA). Time-resolved contrast-enhanced computed tomography (4D CT) captures left atrial (LA) opacification and washout, but it does not directly provide quantitative stasis metrics such as blood residence time. Patient-specific computational fluid dynamics (CFD) can quantify LA/LAA residence time, yet routine clinical use is limited by computational cost and sensitivity to patient-specific boundary conditions. Here, we present two complementary approaches to infer time-resolved 3D residence time fields directly from contrast dynamics. First, a physics-informed neural network (PINN) treats contrast as a passive scalar and jointly reconstructs velocity and residence time by enforcing the incompressible Navier-Stokes equations and transport equations for contrast concentration and residence time in moving, patient-specific LA anatomies. Second, an indicator dilution theory (IDT) formulation computes voxelwise, time-resolved residence time maps from contrast time curves alone by constructing a PV-referenced impulse response and modeling transport with a tank-in-series model with spatially dependent parameters. Both methods are benchmarked against patient-specific CFD in six cases spanning diverse LA function, including three patients with TIA or thrombus in the LAA and three patients free of events. Both approaches reproduce expected spatial and temporal trends, with higher residence time in the distal LAA and higher LAA residence time in cases with TIA or thrombus. IDT demonstrates the closest agreement with CFD across the full range of residence times and produces maps in seconds, facilitating clinical translation. In contrast, the PINN additionally recovers phase-dependent atrial flow structures, but tends to smooth and underestimate the highest residence-time regions and requires hours of training. Together, these results support a scalable workflow in which IDT enables rapid stasis screening from contrast CT, and PINNs provide a complementary pathway for detailed, patient-specific hemodynamic inference when full-field flow information is needed.

  PublisherDOI

• Double trouble: Prenatal CMR for fetal cardiac mass in a twin gestation

  Journal of Cardiovascular Magnetic Resonance · 2025-01-01

  articleOpen access
  PublisherDOI

• Three‐dimensional regional evaluation of right ventricular myocardial work from cine computed tomography: A pilot study

  Medical Physics · 2025-03-19 · 4 citations

  articleOpen accessSenior authorCorresponding

  BACKGROUND: Evaluating regional variations in right ventricular (RV) performance can be challenging, particularly in patients with significant impairments due to the need for 3D spatial coverage with high spatial resolution. ECG-gated cineCT can fully visualize the RV and be used to quantify regional strain with high spatial resolution. However, strain is influenced by loading conditions. Myocardial work (MW)-measured clinically as the ventricular pressure-strain loop area-is considered a more comprehensive metric due to its independence of preload and afterload. In this study, we sought to develop regional RV MW assessments in 3D with high spatial resolution by combining cineCT-derived regional strain with RV pressure waveforms from right heart catheterization (RHC). PURPOSE: Regional MW is not measured in the right ventricle (RV) due to a lack of high spatial resolution regional strain (RS) estimates throughout the ventricle. We present a cineCT-based approach to evaluate regional RV performance and demonstrate its ability to phenotype three complex populations: end-stage LV failure (HF), chronic thromboembolic pulmonary hypertension (CTEPH), and repaired tetralogy of Fallot (rTOF). METHODS: Forty-nine patients (19 HF, 11 CTEPH, 19 rTOF) underwent cineCT and RHC. RS was estimated as the regional change in the endocardial surface from full-cycle ECG-gated cineCT and combined with RHC pressure waveforms to create regional pressure-strain loops; endocardial MW was measured as the loop area. Detailed, 3D mapping of RS and MW enabled spatial visualization of strain and work strength, and phenotyping of patients. RESULTS: HF patients demonstrated more overall impaired strain and work compared to the CTEPH and rTOF cohorts. For example, the HF patients had more akinetic areas (median: 9%) than CTEPH (median: < 1%, p = 0.02) and rTOF (median: 1%, p < 0.01) and performed more low work (median: 69%) than the rTOF cohort (median: 38%, p < 0.01). The CTEPH cohort had more impairment in the septal wall; < 1% of the free wall and 16% of the septal wall performed negative work. The rTOF cohort demonstrated a wide distribution of strain and work, ranging from hypokinetic to hyperkinetic strain and low to medium-high work. Impaired strain (-0.15 ≤ RS) and negative work were strongly-to-very strongly correlated with RVEF (R = -0.89, p < 0.01; R = -0.70, p < 0.01, respectively), while impaired work (MW ≤ 5 mmHg) was moderately correlated with RVEF (R = -0.53, p < 0.01). CONCLUSION: Regional RV MW maps can be derived from clinical CT and RHC studies and can provide patient-specific phenotyping of RV function in complex heart disease patients.

  PublisherOA PDFDOI

• Prenatal fetal CMR for postnatal intervention planning in obstructed infradiaphragmatic total anomalous pulmonary venous return

  Journal of Cardiovascular Magnetic Resonance · 2025-01-01

  articleOpen access
  PublisherDOI

• Deep learning enables fully automated cineCT-based assessment of regional right ventricular function

  medRxiv · 2025-09-30

  preprintOpen accessSenior authorCorresponding

  Abstract Background Right ventricular (RV) function is a key factor in the diagnosis and prognosis of heart disease. However, current advanced CT-based assessments rely on semi-automated segmentation of the RV blood pool and manual delineation of the RV free and septal wall boundaries. Both of these steps are time-consuming and prone to inter- and intra-observer variability. Methods We developed and evaluated a fully automated pipeline consisting of two deep learning methods to automate volumetric and regional strain analysis of the RV from contrast-enhanced, ECG-gated cineCT images. The Right Heart Blood Segmenter (RHBS) is a 3D high resolution configuration of nnU-Net to define the endocardial boundary, while the Right Ventricular Wall Labeler (RVWL) is a 3D point cloud-based deep learning method to label the free and septal walls. We trained our models using a diverse cohort of patients with different RV phenotypes and tested in an independent cohort of patients with aortic stenosis undergoing TAVR. Results Our approach demonstrated high accuracy in both cross-validation and independent validation cohorts. RHBS and RVWL both yielded Dice scores of 0.96, and accurate volumetry metrics. RVWL achieved high Dice scores (&gt;0.90) and high accuracy (&gt;93%) for wall labeling. The combination of RHBS and RVWL provided accurate assessment of free and septal wall regional strain, with a median cosine similarity value of 0.97 in the independent cohort. Conclusions A fully automated 3D cineCT-based RV regional strain analysis pipeline has the potential to significantly enhance the efficiency and reproducibility of RV function assessment, enabling the evaluation of large cohorts and multi-center studies. Key Points RV endocardial segmentation of contrast enhanced CT scans can be utilized to perform volumetry, and when paired with labeling of free and septal walls, regional evaluation of surface strain. However, this has previously been performed using time-intensive semi-automated segmentation methods and manually labeling free wall and septal wall regions.. Here, we describe an automated, deep learning-based approach which uses two separate DL models to define the endocardial boundary (in 3D) and then label the free and septal walls on the endocardial surface. Our approach facilitates rapid and automatic advanced phenotyping of patients. This reduces prior limitations of potential interobserver variability and challenges associated with evaluating large cohorts.

  PublisherOA PDFDOI

• Computed Tomography (CT)-derived Cardiovascular Flow Estimation Using Physics-Informed Neural Networks Improves with Sinogram-based Training: A Simulation Study

  ArXiv.org · 2025-11-05

  preprintOpen accessSenior author

  Background: Non-invasive imaging-based assessment of blood flow plays a critical role in evaluating heart function and structure. Computed Tomography (CT) is a widely-used imaging modality that can robustly evaluate cardiovascular anatomy and function, but direct methods to estimate blood flow velocity from movies of contrast evolution have not been developed. Purpose: This study evaluates the impact of CT imaging on Physics-Informed Neural Networks (PINN)-based flow estimation and proposes an improved framework, SinoFlow, which uses sinogram data directly to estimate blood flow. Methods: We generated pulsatile flow fields in an idealized 2D vessel bifurcation using computational fluid dynamics and simulated CT scans with varying gantry rotation speeds, tube currents, and pulse mode imaging settings. We compared the performance of PINN-based flow estimation using reconstructed images (ImageFlow) to SinoFlow. Results: SinoFlow significantly improved flow estimation performance by avoiding propagating errors introduced by filtered backprojection. SinoFlow was robust across all tested gantry rotation speeds and consistently produced lower mean squared error and velocity errors than ImageFlow. Additionally, SinoFlow was compatible with pulsed-mode imaging and maintained higher accuracy with shorter pulse widths. Conclusions: This study demonstrates the potential of SinoFlow for CT-based flow estimation, providing a more promising approach for non-invasive blood flow assessment. The findings aim to inform future applications of PINNs to CT images and provide a solution for image-based estimation, with reasonable acquisition parameters yielding accurate flow estimates.

  PublisherOA PDFDOI

• Autonomous 5D-flow with Radial k-Space Sampling

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2024-08-14

  articleSenior author

  Autonomous acquisition of radial k-space data improves sample uniformity in 2D cardiac MRI. We aim to extend this approach to 5D-flow imaging (4D flow with respiratory gating). We have simulated data acquisition using an autonomous approach (ARKS) and compared it to golden-angle (GA) based spiral phyllotaxis acquisition. Simulations were based on physiologic data recorded from pediatric patients undergoing conventional 4D flow imaging. We found that a 4.5 min ARKS scan achieves a higher degree of sampling uniformity than an 8 min GA scan. Future work will focus on implementing 5D-flow ARKS in vivo and evaluating the impact on flow accuracy.

  PublisherDOI

Recent grants

• Use of Real Time MRI for Evaluation of Regional Myocardial Function

  NIH · $49k · 2014–2016

• Comprehensive Assessment of Right Ventricular and Pulmonary Vascular Function via CT Imaging in Heart Failure Patients

  NIH · $872k · 2018–2023

Frequent coauthors

• Elliot R. McVeigh

  University of California, San Diego

  78 shared

• Zhennong Chen

  Harvard University

  46 shared

• Walter R. Witschey

  36 shared

• Robert C. Gorman

  University of Nebraska–Lincoln

  34 shared

• Yuchi Han

  29 shared

• Anderson Scott

  La Jolla Bioengineering Institute

  22 shared

• Gabrielle Colvert

  University of California, San Diego

  20 shared

• Amanda Craine

  University of California, San Diego

  19 shared

Labs

• Contijoch Research LaboratoryPI

  Labe member profiles in Prof. Francisco Contijoch's research group

Education

• PhD, Bioengineering

  University of Pennsylvania

  2015

• MSE, Biomedical Engineering

  Johns Hopkins University

  2010

• BS, Biomedical Engineering

  Johns Hopkins University

  2008

Awards & honors

• Howard Hughes Medical Institute (HHMI) – NIH NIBIB Interface…
• National Institutes of Health National Heart, Lung, and Bloo…
• UCSD Frontiers of Innovation Scholars Postdoctoral Fellowshi…
• University of California President’s Postdoctoral Fellowship