Cardiac epicardial and mediastinal fat volumes correlate: the feasibility of predicting one based on the other

• Published in: Computers in biology and medicine
• Main Authors: Rodrigues, É.O, Pinheiro, V.H.A, Liatsis, P, Conci, A
• Format: Journal Article
• Language: English
• Published: 2017
• Subjects: Internal Medicine; Other; Mediastinal; Correlation; Adipose tissue; Quantification; Prediction; Regression; Cardiac fat segmentation; Epicardial
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2017.02.010

Abstract We propose a methodology to predict the cardiac epicardial and mediastinal fat volumes in Computed Tomography images using regression algorithms. We conclude that it is feasible to predict these fats with a high degree of correlation, thus alleviating the requirement for manual or automatic segmentation of both fat volumes. Instead, segmenting just one of them suffices, while the volume of the other may be predicted fairly precisely. The correlation coefficient obtained by the Rotation Forest algorithm using the MLP Regressor in predicting the mediastinal fat based on the epicardial fat is 0.9876, with a relative absolute error of 14.4% and a root relative squared error of 15.7%. The best correlation coefficient obtained in predicting the epicardial fat based on the mediastinal information is 0.9683 with a relative absolute error of 19.6% and a relative squared error of 24.9%. Moreover, we analysed the feasibility of using linear regressors, which provide an intuitive interpretation of the underlying approximations. Specifically, when using linear regression for prediction of the mediastinal fat based on the epicardial fat the correlation coefficient is 0.9534, with a relative absolute error of 31.6% and a root relative squared error of 30.1%. In the case of prediction of the epicardial fat based on the mediastinal fat using the linear regressor, the correlation coefficient is 0.8531, with a relative absolute error of 50.43% and a root relative squared error of 52.06%. Using this approach, it is possible to speed up some segmentation and quantification methods, currently employed in the state-of-the-art, as well as subsequent medical analysis, thus supporting the prevention of health problems and reducing undesirable outcomes.