A novel artificial intelligence (AI) algorithm shows promise for estimating left ventricular ejection function (LVEF) using routinely obtained left coronary artery angiogram videos, a new study suggests.
In the test dataset, the video-based algorithm, called a deep neural network (DNN), discriminated reduced LVEF (<40%) with an area under the receiver operating characteristic curve (AUROC) of 0.911.
In the external validation dataset, the DNN discriminated reduced LVEF with an AUROC of 0.906.
However, the DNN tended to overestimate low LVEFs and to underestimate high LVEFs.
"We know the findings will be unexpected for cardiologists who don't typically expect to get an estimate of systolic function or pump function just from an angiogram," principal investigator Geoffrey H. Tison, MD, of the University of California, San Francisco (UCSF), told theheart.org | Medscape Cardiology.
In fact, he noted, "One of the challenges we face is a lack of trust by the healthcare community. They may not understand what drives the predictions behind our models. We have to translate that information in such a way that physicians trust that the algorithm is using the right features from the data they feed in to make the predictions."
To help bolster that trust, "we display the 'Model Facts,' a nutrition-style label that describes how we train the algorithm, how it was validated, and the inclusion and exclusion criteria," lead author Robert Avram, MD, of the University of Montreal, Canada, added.
Model Facts is a safeguard against inappropriate use of the algorithm, Avram said. For example, if the algorithm was trained on patients between the ages of 40 and 90 and a clinician fed in data for a 35-year-old, a pop-up would appear warning the physician that the data being inputted are different from the data the algorithm was trained and validated on, and so any prediction "should be taken with a grain of salt."
The study was published online May 10 in JAMA Cardiology.
Additional Procedure
LVEF can be determined before coronary angiography with transthoracic echocardiography, but that is not always available, particularly for patients being seen emergently for acute coronary syndromes, the researchers write. LVEF can also be assessed using left ventriculography, an additional procedure that requires insertion of a pigtail catheter into the left ventricle and injection of more contrast and longer radiation exposure.
"Novel methods to assess LVEF at the point of care during coronary angiography would expand the available options to perform this important physiologic determination," they write. "Video-based deep neural networks can learn subtle patterns from medical data to accomplish certain tasks beyond what physicians can achieve with that data, providing an opportunity to assess cardiac systolic function in real time from standard angiographic images without additional cost or procedures."
The investigators conducted a cross-sectional study using patient data from 2012–2019 from UCSF. Data were randomly categorized into training, development, and test datasets.
External validation data were obtained from the University of Ottawa Heart Institute.
All adult patients who received a coronary angiogram and a transthoracic echocardiogram (TTE) within 3 months before or 1 month after receiving the angiogram were included.
A total of 4042 angiograms with corresponding TTE LVEF from 3679 UCSF patients were included in the analysis. The mean age of the patients was 64.3 years, and 65% were men.
The researchers' video-based DNN, called CathEF, was used to discriminate reduced LVEF and to predict a continuous LVEF percentage from standard angiogram videos of the left coronary artery.
In the UCSF test dataset, CathEF discriminated reduced LVEF with an AUROC of 0.911; the diagnostic odds ratio for reduced LVEF was 22.7.
Furthermore, the CathEF-predicted that LVEF had a mean absolute error (MAE) of 8.5% compared with TTE LVEF.
The CathEF-predicted LVEF differed 5% or less in comparison with the TTE LVEF in 38% of the test dataset studies; however, differences greater than 15% were seen in 15.2% of cases.
In the external validation, CathEF discriminated reduced LVEF with an AUROC of 0.906 and an MAE of 7%.
CathEF performance was consistent irrespective of patient characteristics, including sex, body mass index, low estimated glomerular filtration rate (<45), acute coronary syndromes, obstructive coronary artery disease, and left ventricular hypertrophy.
However, as noted, it tended to overestimate low LVEFs and to underestimate high LVEFs.
"Further research can improve accuracy and reduce the variability of DNNs to maximize their clinical utility," the authors conclude.
A validation study is underway at the Montreal Heart Institute, and similar studies are planned at UCSF and McGill University, Tison said. "We expect to present preliminary findings at medical conferences either before the end of the year or maybe for the American College of Cardiology meeting in March 2024."
Potentially Improved Outcomes
Alfonso H. Waller, MD, a member of the American College of Cardiology's Imaging Council and director of cardiac imaging at Rutgers New Jersey Medical School, commented on the study for theheart.org | Medscape Cardiology.
"At some centers, in patients presenting with an acute ST-segment elevation myocardial infarction (STEMI), some argue that assessment of myocardial and valvular function with LV [left ventriculography] grams may provide important prognostic information and in part may help guide the management of the patient," he said.
"Therefore, this novel approach may provide information that is not usually available without performing a classic LV gram...[and] may lead to improved delivery of care, earlier therapies, and potentially improved outcomes and quality of life."
If the technology is available in real time, "it could enable real-time, dynamic assessment of cardiac function during coronary angiography, which may be particularly helpful in acute STEMI cases where baseline cardiac function and renal function may be unknown and additional contrast may be detrimental," he said.
However, patients who might benefit most from the technology are those with severely reduced LVEF, "and unfortunately, the LVEF may be overestimated in this group," he said.
Waller also noted that the model was developed using echocardiograms obtained 3 months before or up to 1 month after the angiogram, during which time "LVEF may change significantly. Typically, if someone presents with an acute coronary syndrome, there can be myocardial stunning, which can lead to regional wall motion abnormalities and lowering of LVEF."
The validation study is evaluating patients with acute coronary syndrome for whom an echocardiogram was performed within 48 hours of the angiogram, he added.
The study was supported by grants from the Fonds de la Recherche en Santé du Québec, the Montreal Heart Institute Research Centre, the Montreal Heart Institute Foundation, the Des Groseillers-Bérard Research Chair, the National Institutes of Health, and the Heart and Stroke Foundation of Ontario. Tison, Avram and Waller have disclosed no relevant financial relationships.
JAMA Cardiology. Published online May 10, 2023. Abstract
Follow Marilynn Larkin on Twitter: @MarilynnL.
For more from the heart.org | Medscape Cardiology, follow us on Twitter and Facebook.
Credits:
Lead image: Pitchayanan Kongkaew/Dreamstime
Medscape Medical News © 2023
Cite this: Video-Based AI Tool Estimates LVEF From Angiograms - Medscape - May 15, 2023.
Comments