Longitudinal Prediction of Infant MR Images With Multi-Contrast Perceptual Adversarial Learning

Abstract

The infant brain undergoes a remarkable period of neural development that is crucial for the development of cognitive and behavioral capacities (Hasegawa et al., 2018). Longitudinal magnetic resonance imaging (MRI) is able to characterize the developmental trajectories and is critical in neuroimaging studies of early brain development. However, missing data at different time points is an unavoidable occurrence in longitudinal studies owing to participant attrition and scan failure. Compared to dropping incomplete data, data imputation is considered a better solution to address such missing data in order to preserve all available samples. In this paper, we adapt generative adversarial networks (GAN) to a new application: longitudinal image prediction of structural MRI in the first year of life. In contrast to existing medical image-to-image translation applications of GANs, where inputs and outputs share a very close anatomical structure, our task is more challenging as brain size, shape and tissue contrast vary significantly between the input data and the predicted data. Several improvements over existing GAN approaches are proposed to address these challenges in our task. To enhance the realism, crispness, and accuracy of the predicted images, we incorporate both a traditional voxel-wise reconstruction loss as well as a perceptual loss term into the adversarial learning scheme. As the differing contrast changes in T1w and T2w MR images in the first year of life, we incorporate multi-contrast images leading to our proposed 3D multi-contrast perceptual adversarial network (MPGAN). Extensive evaluations are performed to assess the qualityand fidelity of the predicted images, including qualitative and quantitative assessments of the image appearance, as well as quantitative assessment on two segmentation tasks. Our experimental results show that our MPGAN is an effective solution for longitudinal MR image data imputation in the infant brain. We further apply our predicted/imputed images to two practical tasks, a regression task and a classification task, in order to highlight the enhanced task-related performance following image imputation. The results show that the model performance in both tasks is improved by including the additional imputed data, demonstrating the usability of the predicted images generated from our approach.

Generative AI models in time varying biomedical data: a systematic review (Preprint)

“…GP-GAN uses stacked GANs to predict tumor growth from longitudinal MRI [105]. MPGAN makes longitudinal Prediction of Infant MRI With Multi-Contrast Perceptual Adversarial Learning [106]. In LD-GAN [107], the authors predict disease progression images with missing MRI in the input.…”

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

“…sMRI is faster and contains infant-specific procedures, such as BCP ( Howell et al, 2019 ; Gao K. et al, 2021 ). Deep generative algorithms were first utilized to predict ASD in infants using longitudinal data by Peng and others ( Peng et al, 2021 ). Hazlett et al (2017) developed a three-stage SAE model to diagnose infants with autism before the onset of behavioral signs.…”

Translating the future: image-to-image translation for the prediction of future brain metabolism

“…Two previous studies employed similar network architectures for the prediction of longitudinal brain changes. Peng et al 8 predicted brain changes on MRI in infants using a generative adversarial network, where the generator, i.e., the part of the network that predicted the images, was based on a supervised convolutional neural network (CNN). In the only currently published approach for the longitudinal prediction of [ 18 F]-FDG PET, Choi et al 9 implemented an unsupervised conditional variational autoencoder, which generated realistic follow-up data in a small sample (n = 26) of cognitively normal (CN) subjects based on baseline data and age information.…”