Prediction of prognosis in patients with tetralogy of Fallot based on deep learning imaging analysis

Categories:

Prediction of prognosis in patients with tetralogy of Fallot based on deep learning imaging analysis

Diller GP, Orwat S, Vahle J, Bauer UMM, Urban A, Sarikouch S, Berger F, Beerbaum P, Baumgartner H; German Competence Network for Congenital Heart Defects Investigators.

Heart. 2020 Mar 11. pii: heartjnl-2019-315962. doi: 10.1136/heartjnl-2019-315962. [Epub ahead of print]

PMID: 32161041

Similar articles

Select item 32159755
 
Take Home Points

  • Machine learning algorithms applied to cardiac magnetic resonance (CMR) images can automatically estimate prognosis in patients with repaired tetralogy of Fallot (ToF)
  • The current study highlights the prognostic value of automatically derived right atrial area and biventricular dysfunction in patients with ToF.
  • Enlarged right atrial median area (HR 1.11/ cm², p=0.003) and reduced right ventricular long-axis strain (HR 0.80/%, p=0.009) was associated with adverse endpoint of death/aborted cardiac arrest or documented ventricular tachycardia (defined as >3 documented consecutive ventricular beats).
  • Machine learning algorithms trained on external imaging datasets can automatically estimate prognosis in patients with ToF.

 
Dr. Shaji Menon (Salt Lake City, Utah)
 
Comment from Dr. Shaji Menon (Salt Lake City, Utah), Lead Section editor of Pediatric Cardiology Journal Watch: This study assesses the utility of machine learning algorithms for automatically estimating prognosis in patients with repaired tetralogy of Fallot (ToF) using Cardiac magnetic resonance (CMR). The study included 372 patients with ToF who had undergone CMR imaging as part of German National Registry for Congenital Heart Disease between 2003 and 2009. Cine loops were retrieved and subjected to automatic deep learning (DL)-based image analysis, trained on independent, local CMR data, to derive measures of cardiac dimensions and function. This information was combined with established clinical parameters and ECG markers of prognosis. Over a median follow-up period of 10 years, 23 patients experienced an endpoint of death/aborted cardiac arrest or documented ventricular tachycardia (defined as >3 documented consecutive ventricular beats). On univariate Cox analysis, various DL parameters, including right atrial median area (HR 1.11/ cm², p=0.003) and right ventricular long-axis strain (HR 0.80/%, p=0.009) emerged as significant predictors of outcome. DL parameters were related to adverse outcome independently of left and right ventricular ejection fraction and peak oxygen uptake (p<0.05 for all). A composite score of enlarged right atrial area and depressed right ventricular longitudinal function identified a ToF subgroup at significantly increased risk of adverse outcome (HR 2.1/unit, p=0.007). Risk stratification models for patients with ToF have traditionally included a variety of parameters, including age at repair, duration of preoperative cyanosis, previous atrial arrhythmias, QRS duration or fragmentation on ECG, biventricular dysfunction, right atrial area, RV hypertrophy, elevated left ventricular end-diastolic pressure and natriuretic peptides. This study for the first time highlights the prognostic value of automatically derived right atrial area and biventricular dysfunction in patients with ToF.
 
Study Overview
 
Demographic and clinical characteristics
 
Cardiac MRI Parameters
 
Illustration of Segmentation Results
 Results of the univariate Cox proportional hazard analysis
 
Cox analysis-based survivor function
 
 

Atarim

Dive straight into the feedback!
Login below and you can start commenting using your own user instantly