Gp-Unet
Lesion detection from weak labels with a 3D regression network
Florian Dubost (Erasmus MC)
Gerda Bortsova (Erasmus MC)
Hieab H. Adams (Erasmus MC)
M. Arfan Ikram (Erasmus MC)
Wiro J. Niessen (Erasmus MC, TU Delft - Applied Sciences, TU Delft - ImPhys/Quantitative Imaging)
Meike W. Vernooij (Erasmus MC)
Marleen de Bruijne (Erasmus MC, University of Copenhagen)
More Info
expand_more
Abstract
We propose a novel convolutional neural network for lesion detection from weak labels. Only a single, global label per image - the lesion count - is needed for training. We train a regression network with a fully convolutional architecture combined with a global pooling layer to aggregate the 3D output into a scalar indicating the lesion count. When testing on unseen images, we first run the network to estimate the number of lesions. Then we remove the global pooling layer to compute localization maps of the size of the input image. We evaluate the proposed network on the detection of enlarged perivascular spaces in the basal ganglia in MRI. Our method achieves a sensitivity of 62% with on average 1.5 false positives per image. Compared with four other approaches based on intensity thresholding, saliency and class maps, our method has a 20% higher sensitivity.