February 14, 2023Open Access

Multi-center retrospective cohort study applying deep learning to electrocardiograms to identify left heart valvular dysfunction

Q: What is the clinical evidence from this study?

Study design: Cohort. Population: Aortic Stenosis and Mitral Regurgitation (n=128628). Intervention: Deep learning model applied to electrocardiograms vs. Echocardiogram (ground truth). Primary outcome: Detection of Aortic Stenosis (AUROC 0.89, 95% CI 0.88-0.89).

AVAkhil VaidCross-Cutting Cardiology EAEdgar ArgulianMount Sinai Hospital SLStamatios LerakisCardiac Imaging

Resultado clave

A deep learning model applied to electrocardiograms detected aortic stenosis with an AUROC of 0.89 and mitral regurgitation with an AUROC of 0.88 in internal testing.

Diseño del estudio

Tipo

Cohort (n=128,628)

Multicéntrico

Sí

PICO estructurado

Does a deep learning model applied to electrocardiograms identify moderate-to-severe left heart valvular dysfunction in a diverse patient cohort?

Población

128,628 patients with 617,338 Transthoracic Echocardiogram-Electrocardiogram (Echo-ECG) pairs from five Mount Sinai hospitals within New York City, representing a demographically diverse cohort.

Intervención

Deep learning model applied to standard 12-lead electrocardiograms (ECGs) to detect moderate-to-severe or severe Mitral Regurgitation (MR) and Aortic Stenosis (AS).

Resultado

Detection of moderate-to-severe or severe Mitral Regurgitation and Aortic Stenosis, measured by Area Under Receiver Operating Characteristic curve (AUROC).surrogate

Deep learning algorithms applied to standard ECGs can accurately detect moderate-to-severe aortic stenosis and mitral regurgitation, offering a potential low-cost, widely available screening tool for valvular heart disease.

Resultado numérico

Estimación del efecto: AUROC 0.89 (95% CI 0.88-0.89)

Limitaciones

Echocardiograms were paired to ECGs within a ±7-day period, which may decrease accuracy for acute severe mitral regurgitation.
Random error may have been introduced by the labels generated by the NLP pipeline.
Low outcome prevalence in the study population led to proportionally low Positive Predictive Values (PPV).
External validation was performed at a hospital within the same health system, warranting further validation from another health system.

Resumen

BACKGROUND: Aortic Stenosis and Mitral Regurgitation are common valvular conditions representing a hidden burden of disease within the population. The aim of this study was to develop and validate deep learning-based screening and diagnostic tools that can help guide clinical decision making. METHODS: In this multi-center retrospective cohort study, we acquired Transthoracic Echocardiogram reports from five Mount Sinai hospitals within New York City representing a demographically diverse cohort of patients. We developed a Natural Language Processing pipeline to extract ground-truth labels about valvular status and paired these to Electrocardiograms (ECGs). We developed and externally validated deep learning models capable of detecting valvular disease, in addition to considering scenarios of clinical deployment. RESULTS: We use 617,338 ECGs paired to transthoracic echocardiograms from 123,096 patients to develop a deep learning model for detection of Mitral Regurgitation. Area Under Receiver Operating Characteristic curve (AUROC) is 0.88 (95% CI:0.88-0.89) in internal testing, and 0.81 (95% CI:0.80-0.82) in external validation. To develop a model for detection of Aortic Stenosis, we use 617,338 Echo-ECG pairs for 128,628 patients. AUROC is 0.89 (95% CI: 0.88-0.89) in internal testing, going to 0.86 (95% CI: 0.85-0.87) in external validation. The model's performance increases leading up to the time of the diagnostic echo, and it performs well in validation against requirement of Transcatheter Aortic Valve Replacement procedures. CONCLUSIONS: Deep learning based tools can increase the amount of information extracted from ubiquitous investigations such as the ECG. Such tools are inexpensive, can help in earlier disease detection, and potentially improve prognosis.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo