Automatic Prediction of Paediatric Cardiac Output From Echocardiograms Using Deep Learning Models

Ufkes, Steven; Zuercher, Mael; Erdman, Lauren; Slorach, Cameron; Mertens, Luc; Taylor, Katherine

doi:10.1016/j.cjcpc.2022.11.001

Cited by 2 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research in this area includes the evaluation of heart sounds with artificial neural networks, classification of risk for pediatric cardiac surgery from pre‐ and post‐surgical data and inpatient laboratory values for prediction of critical events in infants with single‐ventricle physiology 12–14 . The EchoNet‐Dynamic model, trained of over 10,000 adult echocardiograms, was retrained on pediatric echocardiograms and successfully used to identify reduced cardiac output 15 …”

Section: Artificial Intelligence In Congenital Heart Diseasementioning

confidence: 99%

“…[12][13][14] The EchoNet-Dynamic model, trained of over 10,000 adult echocardiograms, was retrained on pediatric echocardiograms and successfully used to identify reduced cardiac output. 15 More specific to PDA, Lei et al demonstrated proof-of-concept using deep learning for automated detection of PDA on echocardiograms. 16 While groundbreaking, the sensitivity of their model (76%) would likely need improvement to be acceptable in a clinical tool.…”

Section: Artificial Intelligence In Congenital Heart Diseasementioning

confidence: 99%

See 1 more Smart Citation

Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model

Erno

Gomes

Baltimore

et al. 2023

J of Ultrasound Medicine

View full text Add to dashboard Cite

ObjectivesPatent ductus arteriosus (PDA) is a vascular defect common in preterm infants and often requires treatment to avoid associated long‐term morbidities. Echocardiography is the primary tool used to diagnose and monitor PDA. We trained a deep learning model to identify PDA presence in relevant echocardiographic images.MethodsEchocardiography video clips (n = 2527) in preterm infants were reviewed by a pediatric cardiologist and those relevant to PDA diagnosis were selected and labeled (PDA present/absent/indeterminate). We trained a convolutional neural network to classify each echocardiography frame of a clip as belonging to clips with or without PDA. A novel attention mechanism that aggregated predictions for all frames in each clip to obtain a clip‐level prediction by weighting relevant frames.ResultsIn early model iterations, we discovered training with color Doppler echocardiography clips produced the best performing classifier. For model training and validation, 1145 such clips from 66 patients (661 PDA+ clips, 484 PDA− clips) were used. Our best classifier for clip level performance obtained sensitivity of 0.80 (0.83–0.90), specificity of 0.77 (0.62–0.92) and AUC of 0.86 (0.83–0.90). Study level performance obtained sensitivity of 0.83 (0.72–0.94), specificity of 0.89 (0.79–1.0) and AUC of 0.93 (0.89–0.98).ConclusionsOur novel deep learning model demonstrated strong performance in classifying echocardiography clips with and without PDA. Further model development and external validation are warranted. Ultimately, integration of such a classifier into auto detection software could streamline PDA imaging workflow. This work is the first step toward semi‐automated, bedside detection of PDA in preterm infants.

show abstract

Section: Artificial Intelligence In Congenital Heart Diseasementioning

confidence: 99%