Fig. 1From: Enhancing magnetic resonance imaging-driven Alzheimer’s disease classification performance using generative adversarial learningSchematic of the overall deep learning strategy. The generative adversarial network (GAN) uses 1.5-T and 3-T scans of the same individual taken at the same time to generate images (3 T*). The fully convolutional network (FCN) model uses the 3-T* images to predict Alzheimer’s disease (AD) status. Both the GAN and FCN models were trained simultaneously by backpropagating the losses from the GAN and the FCN modelsBack to article page