A study of CNN capacity applied to Left Venticle Segmentation in Cardiac MRI
This work addresses practical deployment questions for CNNs in medical imaging, but it is incremental as it applies existing methods to analyze capacity without introducing new techniques.
The study investigated how dataset size and network depth affect CNN performance for left ventricle segmentation in cardiac MRI, finding that sample size impacts performance more than architecture or hyperparameters, with small samples being more sensitive to hyperparameters.
CNN (Convolutional Neural Network) models have been successfully used for segmentation of the left ventricle (LV) in cardiac MRI (Magnetic Resonance Imaging), providing clinical measurements. In practice, two questions arise with deployment of CNNs: 1) when is it better to use a shallow model instead of a deeper one? 2) how the size of a dataset might change the network performance? We propose a framework to answer them, by experimenting with deep and shallow versions of three U-Net families, trained from scratch in six subsets varying from 100 to 10,000 images, different network sizes, learning rates and regularization values. 1620 models were evaluated using 5-fold cross-validation by loss and DICE. The results indicate that: sample size affects performance more than architecture or hyper-parameters; in small samples the performance is more sensitive to hyper-parameters than architecture; the performance difference between shallow and deeper networks is not the same across families.