Donate

Science

Our mission is to use FORCE data to improve outcomes for Fontan patients.

Learn more about our published and in progress research below.

Published Studies

High-Performing Fontan Patients

A Fontan Outcome Registry by Cardiac Magnetic Resonance Imaging Study

Tarek Alsaied, MD, MSC, • Runjia Li, MS, • Adam B. Christopher, MD, • Mark A. Fogel, MD, • Timothy C. Slesnick, MD, • Rajesh Krishnamurthy, MD, • Vivek Muthurangu, MD, PHD, • Adam L. Dorfman, MD, • Christopher Z. Lam, MD, • Justin D. Weigand, MD, • Joshua D. Robinson, MD, • Rachael Cordina, MD, • Laura J. Olivieri, MD, • Rahul H. Rathod, MD, • the FORCE Investigators.

BACKGROUND Fontan patients exhibit decreased exercise capacity. However, there is a subset of high-performing Fontan (HPF) patients with excellent exercise capacity.

OBJECTIVES This study aims to: 1) create a Fontan-specific percent predicted peak VO2 tool using exercise data; 2) examine clinical factors associated with HPF patients; and 3) examine late outcomes in HPF patients.

METHODS Patients in the multi-institutional Fontan Outcomes Registry Using CMR Examination above the age of 8 years who had a maximal exercise test were included. An HPF patient was defined as a patient in the upper Fontanspecific percent predicted peak VO2 quartile. Multivariable logistic regression was employed to investigate factors associated with the HPF and Cox regression was used to examine the association between HPF patients and late outcomes (composite of death or listing for cardiac transplant).

RESULTS The study included 813 patients (mean age: 20.2 8.7 years). An HPF patient was associated with left ventricular morphology (OR: 1.50, P . 0.04), mixed morphology (OR: 2.23, P < 0.001), and a higher ejection fraction (OR: 1.31 for 10% increase, P . 0.01). Patients with at least moderate atrioventricular valve regurgitation, protein-losing enteropathy, or who were using psychiatric medications, were less likely to be an HPF patient. After a mean follow-up of 3.7 years, 46 (5.7%) patients developed a composite endpoint. HPF had a lower risk of death or listing for cardiac transplant (HR: 0.06 [95% CI: 0.01–0.25]).

CONCLUSIONS Patients with HPF have more favorable outcomes when compared to patients with lower exercise capacity. This large registry data highlights the role of exercise testing in providing personalized care and surveillance post-Fontan. (JACC Adv. 2024;3:101254) © 2024 The Authors. Published by Elsevier on behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Learn more

A Deep Learning Pipeline for Assessing Ventricular Volumes from a Cardiac MRI Registry of Patients with Single Ventricle Physiology

Tina Yao, MRes* • Nicole St. Clair, BSc* • Gabriel F. Miller, MSc • Adam L. Dorfman, MD • Mark A. Fogel, MD •Sunil Ghelani, MD • Rajesh Krishnamurthy, MD • Christopher Z. Lam, MD • Michael Quail, MD •Joshua D. Robinson, MD • David Schidlow, MD, MMus • Timothy C. Slesnick, MD • Justin Weigand, MD • Jennifer A. Steeden, PhD • Rahul H. Rathod, MD, MBA • Vivek Muthurangu, MD(res)

Abstract: An end-to-end deep learning pipeline was developed to provide automatic segmentation and cardiac function metrics for a cardiac MRI registry of patients with single ventricle physiology; the pipeline requires no human input and is the first to segment images in this patient population.

Purpose: To develop an end-to-end deep learning (DL) pipeline for automated ventricular segmentation of cardiac MRI data from a multicenter registry of patients with Fontan circulation (Fontan Outcomes Registry Using CMR Examinations [FORCE])..

Materials and Methods: This retrospective study used 250 cardiac MRI examinations (November 2007–December 2022) from 13 institutions for training, validation, and testing. The pipeline contained three DL models: a classifier to identify short-axis cine stacks and two U-Net 3+ models for image cropping and segmentation. The automated segmentations were evaluated on the test set (n = 50) by using the Dice score. Volumetric and functional metrics derived from DL and ground truth manual segmentations were evaluated on the test set (n = 50) by using the Dice score. Volumetric and functional metrics derived from DL and ground truth manual segmentations were compared using Bland-Altman and intraclass correlation analysis. The pipeline was further qualitatively evaluated on 475 unseen examinations.ntations were compared using Bland-Altman and intraclass correlation analysis. The pipeline was further qualitatively evaluated on 475 unseen examinations.

Results: There were acceptable limits of agreement (LOA) and minimal biases between the ground truth and DL end-diastolic volume (EDV) (bias: −0.6 mL/m2, LOA: −20.6 to 19.5 mL/m2) and end-systolic volume (ESV) (bias: −1.1 mL/m2, LOA: −18.1 to 15.9 mL/m2), with high intraclass correlation coefficients (ICCs > 0.97) and Dice scores (EDV, 0.91 and ESV, 0.86). There was moderate agreement for ventricular mass (bias: −1.9 g/m2, LOA: −17.3 to 13.5 g/m2) and an ICC of 0.94. There was also acceptable agreement for stroke volume (bias: 0.6 mL/m2, LOA: −17.2 to 18.3 mL/m2) and ejection fraction (bias: 0.6%, LOA: −12.2% to 13.4%), with high ICCs (>0.81). The pipeline achieved satisfactory segmentation in 68% of the 475 unseen examinations, while 26% needed minor adjustments, 5% needed major adjustments, 5% needed major adjustments, and in 0.4%, the cropping model failed.5% needed major adjustments, and in 0.4%, the cropping model failed. and in 0.4%, the cropping model failed.

Conclusion: The DL pipeline can provide fast standardized segmentation for patients with single ventricle physiology across multiple centers. This pipeline can be applied to all cardiac MRI examinations in the FORCE registry.

Learn more

Studies in Progress

Using hierarchical clustering machine learning to discover new subgroups in single ventricle heart patients, enhancing treatment strategies.
Establishing standard CMR imaging results for Fontan patients to improve diagnosis and treatment.
Developing automated tools to analyze 4D flow MRI data, aiding early detection of Fontan failure.
Exploring how weight affects heart characteristics and function in adults with Fontan circulation.
Studying blood flow efficiency in the Fontan circuit to predict long-term health outcomes.
Investigating factors behind 'supranormal' exercise capacity in some Fontan patients to improve overall patient care.

Exploring novel cardiac and hepatic fibrosis biomarkers for Fontan-associated liver disease.
Analyzing ventricular performance and blood flow to anticipate health outcomes in Fontan patients.
Methods paper describing the innovations required to launch a AI-enabled registry so that others can replicate the success of FORCE.
Utilizing machine learning to automate and standardize ventricular size and function measurements in CMR exams.
Investigating the impact of ventricular interaction on systemic RV dysfunction in HLHS patients post-Fontan.
Using MRI and simulations to study heart shape and function in hypoplastic left heart syndrome.

Developing a risk score for sudden cardiac death in Fontan patients to guide clinical decisions.
Applying machine learning to segment Fontan baffles and pulmonary arteries for improved outcome prediction.
Examining the impact of aortic reconstruction on heart performance in Fontan patients using 4D imaging.
Identifying factors predicting diastolic dysfunction in Fontan patients to enhance clinical management.
Creating a comprehensive model to predict death/transplant risk in patients post-Fontan operation.
Using AI to identify early predictors of cardiac complications in single ventricle patients post-Fontan.

This study will compare how the modern style Fontan operations (lateral tunnel versus extracardiac conduit) impact outcomes.
This project investigates if stroke work, a cardiac MRI-derived measure of heart pump function, can predict exercise capacity and complications in patients with Fontan circulation.
This study aims to assess the hemodynamic outcomes of a novel iteration of the Fontan operation. We will analyze the effects of this innovative technique to determine its impact on heart function and overall patient health.
This study explores the long-term impact of creating a small hole (fenestration) during the Fontan procedure. While fenestration helps reduce pressure and improves short-term recovery, its long-term effects are unclear. The goal is to inform better decision-making and management of the fenestration in older Fontan patients.
This study investigates the impact of atrioventricular valve regurgitation on heart function and clinical outcomes in single ventricle patients after the Fontan procedure. We aim to identify risk factors for valve failure by analyzing clinical data, heart imaging, and exercise tests.