A framework of deep learning networks provides expert-level accuracy for the detection and prognostication of pulmonary arterial hypertension.
Gerhard-Paul DillerMaria Luisa Benesch VidalAleksander KempnyKana KubotaWei LiKonstantinos DimopoulosAlexandra ArvanitakiAstrid E LammersStephen J WortHelmut BaumgartnerStefan OrwatMichael A GatzoulisPublished in: European heart journal. Cardiovascular Imaging (2022)
The study highlights the utility of DL algorithms in detecting PAH on routine echocardiograms irrespective of RV dilatation. The algorithms outperform conventional echocardiographic evaluation and provide prognostic information at expert-level. Therefore, DL methods may allow for improved screening and optimized management of PAH.
Keyphrases
- pulmonary arterial hypertension
- deep learning
- machine learning
- pulmonary hypertension
- clinical practice
- pulmonary artery
- mycobacterium tuberculosis
- artificial intelligence
- convolutional neural network
- polycyclic aromatic hydrocarbons
- left ventricular
- left atrial
- social media
- loop mediated isothermal amplification
- atrial fibrillation
- coronary artery
- ejection fraction
- sensitive detection