A deep learning algorithm for VHD diagnosis and evaluation

This study aims to develop and validate a deep learning-based automatic chest radiograph (CXR) cardiovascular border (CB) analysis algorithm (CB_auto) in order to diagnose and quantitatively evaluate valvular heart disease (VHD). The authors found that the CB_auto system, in coordination with the deep learning algorithm, provided highly reliable CB measurements, which, in turn, can be useful, not long daily clinical practice, but also for the purposes of research. Key points A deep learning-based automatic CB analysis algorithm for diagnosing and quantitatively evaluating VHD using posterior-anterior chest radiographs was developed and validated. Our algorithm (CB_auto) yielded comparable reliability to manual CB drawing (CB_hand) in terms of various CB measurement variables, as confirmed by external validation with datasets from three different hospitals and a public dataset. All CB parameters were significantly different between VHD and normal control measurements, and echocardiographic measurements were significantly correlated with CB parameters measured from normal control and VHD CXRs. Article: A deep learning-based automatic analysis of cardiovascular borders on chest radiographs of valvular heart disease: development/external validation Authors: Cherry Kim, Gaeun Lee, Hongmin Oh, Gyujun Jeong, Sun Won Kim, Eun Ju Chun, Young-Hak Kim, June-Goo Lee & Dong Hyun Yang