Rostami, Behrouz

Objective Data on cardiac index (CI) in patients undergoing coronary angiography (CAG) can enhance decision-making and improve clinical outcomes. However, this necessitates an additional invasive procedure, and therefore, it is frequently omitted. This study explores the novel application of convolutional neural networks (CNNs) to predict CI from routine CAGs. Methods and analysis Patients who underwent CAG and simultaneous right heart catheterisation at Mayo Clinic between 2002 and 2023 were included. A three-dimensional (3D)-CNN model (X3D architecture) was developed to predict binary CI categories (normal: CI≥2.2 vs abnormal: CI<2.2 L/min/m²) from CAG videos. CAGs were randomly split into training (70%), validation (15%) and test (15%) sets. All available videos from each study were entered into the model. The model’s performance was evaluated using the area under the curve (AUC), sensitivity, specificity and F1 score. Results A total of 15 297 CAG studies were considered. After applying a quality-check algorithm to exclude low-quality or irrelevant CAGs, 11 475 studies from 9103 patients were included. CI was abnormal in 25.2% of studies. The model achieved an AUC of 0.83 (95% CI 0.81 to 0.85), sensitivity of 0.72 (95% CI 0.68 to 0.77), specificity of 0.78 (95% CI 0.75 to 0.80) and F1 score of 0.62 (95% CI 0.58 to 0.65) on the test dataset. Limiting the input to 4, 3 or 2 prespecified CAG videos representing standard views did not significantly affect the model’s performance, supporting its applicability in varied clinical workflows. Conclusions This study demonstrates the feasibility of leveraging deep learning models to predict CI from CAG videos. The proposed 3D-CNN model shows promise as a tool for providing physicians with real-time data to aid decision-making during CAG procedures.

Abstract Aims Coronary angiography might contain clinically relevant information, beyond its traditional role in delineating coronary artery disease. We sought to explore the role of deep learning (DL) in detecting right ventricular dysfunction from routinely acquired cine images during coronary angiography. Methods and results 3D-convolutional neural networks (CNN) were trained to identify right ventricle (RV) dysfunction from diagnostic angiography of the right coronary artery (RCA). The model’s input was two cine angiograms from LAO and RAO projections. The model’s output was normal vs. abnormal RV function. Ground truth on RV function was obtained from the nearest transthoracic echocardiogram to the coronary angiogram. Two analyses were performed to identify (i) any RV dysfunction (≥mild) and (ii) significant RV dysfunction (≥mild–moderate). A total of 10 336 coronary angiograms from 9849 patients (mean age 66, 36% women) were included. The cohort was split into training (70%), validation (15%), and testing (15%) sets. In the testing sets, the models attained an area under the ROC curve (AUC), sensitivity, and specificity of 0.82 (95% CI: 0.80–0.84), 0.75, and 0.74, for detecting any RV dysfunction, and 0.83 (95% CI: 0.80–0.86), 0.82, and 0.70 for detecting significant (≥mild–moderate) RV dysfunction. Combining an ECG-driven AI model with the angio-driven model improved the AUC values to 0.83 and 0.87 for detecting any or advanced RV dysfunction, respectively. Conclusion A novel deep learning algorithm yielded acceptable classification accuracy in detecting RV dysfunction from routine cine angiography of the RCA. The model’s performance improved further by adding ECG to its inputs.

Abstract Aims Augmenting echocardiography with artificial intelligence would allow for automated assessment of routine parameters and identification of disease patterns not easily recognized otherwise. View classification is an essential first step before deep learning can be applied to the echocardiogram. Methods and results We trained two- and three-dimensional convolutional neural networks (CNNs) using transthoracic echocardiographic (TTE) studies obtained from 909 patients to classify nine view categories (10 269 videos). Transthoracic echocardiographic studies from 229 patients were used in internal validation (2582 videos). Convolutional neural networks were tested on 100 patients with comprehensive TTE studies (where the two examples chosen by CNNs as most likely to represent a view were evaluated) and 408 patients with five view categories obtained via point-of-care ultrasound (POCUS). The overall accuracy of the two-dimensional CNN was 96.8%, and the averaged area under the curve (AUC) was 0.997 on the comprehensive TTE testing set; these numbers were 98.4% and 0.998, respectively, on the POCUS set. For the three-dimensional CNN, the accuracy and AUC were 96.3% and 0.998 for full TTE studies and 95.0% and 0.996 on POCUS videos, respectively. The positive predictive value, which defined correctly identified predicted views, was higher with two-dimensional rather than three-dimensional networks, exceeding 93% in apical, short-axis aortic valve, and parasternal long-axis left ventricle views. Conclusion An automated view classifier utilizing CNNs was able to classify cardiac views obtained using TTE and POCUS with high accuracy. The view classifier will facilitate the application of deep learning to echocardiography.

Background: Enlargement or dilation of the left ventricle (LV) is associated with increased cardiovascular morbidity and mortality, and is an established precursor of ventricular dysfunction and heart failure in almost all major cardiovascular conditions. Although there are existing artificial intelligence frameworks that make use of deep learning to detect LV enlargement, many entail LV segmentation, which can be error-prone especially for point-of-care ultrasound (POCUS) images. Goal: To develop a fully end-to-end deep learning framework that takes as input either transthoracic echocardiographic (TTE) or POCUS 2D image clips, and classifies directly (i.e., without first segmenting the LV) whether a patient has an enlarged LV. Methods: Using a sample of retrospective TTE data enriched for LV enlargement (out of 13,868 patients total, 3,351 had an enlarged LV defined using sex-specific guideline-directed criteria), we trained a classifier that takes as input apical 2-chamber and parasternal long axis videos, and provides as output a binary label denoting whether a patient has an enlarged LV. Then, we evaluated the trained model using a holdout testing dataset of retrospective TTE data (2,087 patients) as well as data from a prospective cohort of 428 patients from whom TTE and POCUS were collected simultaneously. Results: For the retrospective TTE testing dataset, the AUC was 0.851 (95% confidence interval (CI) of 0.830-871). For the prospective cohort, the model performed comparably for both TTE and POCUS video clips (TTE: AUC of 0.865, 95% CI of 0.823-0.906; POCUS: AUC of 0.840, 95% CI of 0.797-0.883). Conclusions: The current deep learning model performed comparably well for both TTE and POCUS echocardiograms. We regard these results as a proof-of-concept of the potential utility of this model in helping point-of-care clinicians make decisions regarding whether patients should receive further evaluation for LV enlargement and associated conditions.

Background: Recently, our team developed an end-to-end artificial intelligence (AI) framework that can successfully estimate left ventricular ejection fraction (LVEF) from echocardiogram videos. However, this framework requires specific views (A2C and PLAX), and cannot estimate LVEF for studies from which these views are missing. Goal: To develop a deep learning model to estimate LVEF from any type and number of echocardiogram video(s). Methods: We built a deep learning model for B-mode echocardiogram videos of the left ventricle (A2C, A3C, A4C, or PLAX views), using retrospective transthoracic echocardiography (TTE) data. The model was evaluated with two test datasets: one with random valid (i.e., showing the left ventricle) views, and one with preselected views (both having 433 patients). Later, the model was tested with a prospective patient cohort that had both TTE and point-of-care ultrasound (POCUS) data collected simultaneously (393 patients). Results: On the retrospective TTE test set, we observed excellent performance (five random valid views per study: RMSE of 6.87% and Pearson correlation (PC) of 0.82 on LVEF estimation, and an AUC of 0.96 (95% CI: of 0.94-0.98) on low LVEF classification (<=40%); preselected views: RMSE of 6.75%, PC of 0.83, and an AUC of 0.96 (95% CI: 0.94-0.98)). For both TTE and POCUS data from the prospective cohort, we also observed strong performance (TTE: RMSE of 5.62%, PC of 0.87, and an AUC of 0.97; POCUS: RMSE of 6.12%, PC of 0.84, and an AUC of 0.96). For the prospective test sets as well as for the retrospective TTE test set with preselected views, results reflect the estimated LVEF average of A2C, A3C, A4C, and PLAX videos from each patient. Conclusions: For both TTE and POCUS data, the current model showed strong performance which was comparable to that of our previously established view-specific model. Alongside this comparable performance, the current model surpasses our previous model in terms of flexibility, as it isn't dependent on view classification and allows a broader selection in terms of the number and types of videos per study or patient. This innovative approach could prove particularly beneficial in point-of-care situations in which it may not be feasible to obtain high-quality, view-specific videos.