Open AccessMulticenter Validation of Convolutional Neural Networks for Automated Detection of Cardiomegaly on Chest Radiographs
Author(s) -
Diego J. Cárdenas,
José Ferreira,
Ramón Moreno,
M.S. Rebelo,
José Eduardo Krieger,
Marco Antônio Gutierrez
Publication year - 2020
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/sbcas.2020.11512
Subject(s) - convolutional neural network , radiography , computer science , artificial intelligence , generalization , multilayer perceptron , artificial neural network , test set , pattern recognition (psychology) , cross validation , set (abstract data type) , radiology , medicine , mathematics , mathematical analysis , programming language
This work focused on validating five convolutional neural network models to detect automatically cardiomegaly, a health complication that causes heart enlargement, which may lead to cardiac arrest. To do that, we trained the models with a customized multilayer perceptron. Radiographs from two public datasets were used in experiments, one of them only for external validation. Images were pre-processed to contain just the chest cavity. The EfficientNet model yielded the highest area under the curve (AUC) of 0.91 on the test set. However, the Inception-based model obtained the best generalization performance with AUC of 0.88 on the independent multicentric dataset. Therefore, this work accurately validated radiographic models to identify patients with cardiomegaly.