Pooling MRI data from multiple datasets requires harmonization to reduce undesired inter-site variabilities, while preserving effects of biological variables (or covariates). The popular harmonization approach ComBat uses a mixed effect regression framework that explicitly accounts for covariate distribution differences across datasets. There is also significant interest in developing harmonization approaches based on deep neural networks (DNNs), such as conditional variational autoencoder (cVAE). However, current DNN approaches do not explicitly account for covariate distribution differences across datasets. Here, we provide mathematical results, suggesting that not accounting for covariates can lead to suboptimal harmonization. We propose two DNN-based covariate-aware harmonization approaches: covariate VAE (coVAE) and DeepResBat. The coVAE approach is a natural extension of cVAE by concatenating covariates and site information with site- and covariate-invariant latent representations. DeepResBat adopts a residual framework inspired by ComBat. DeepResBat first removes the effects of covariates with nonlinear regression trees, followed by eliminating site differences with cVAE. Finally, covariate effects are added back to the harmonized residuals. Using three datasets from three continents with a total of 2787 participants and 10,085 anatomical T1 scans, we find that DeepResBat and coVAE outperformed ComBat, CovBat and cVAE in terms of removing dataset differences, while enhancing biological effects of interest. However, coVAE hallucinates spurious associations between anatomical MRI and covariates even when no association exists. Future studies proposing DNN-based harmonization approaches should be aware of this false positive pitfall. Overall, our results suggest that DeepResBat is an effective deep learning alternative to ComBat. Code for DeepResBat can be found here: https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/harmonization/An2024_DeepResBat.

We introduce *Brain-JEPA*, a brain dynamics foundation model with the Joint-Embedding Predictive Architecture (JEPA). This pioneering model achieves state-of-the-art performance in demographic prediction, disease diagnosis/prognosis, and trait prediction through fine-tuning. Furthermore, it excels in off-the-shelf evaluations (e.g., linear probing) and demonstrates superior generalizability across different ethnic groups, surpassing the previous large model for brain activity significantly. Brain-JEPA incorporates two innovative techniques: **Brain Gradient Positioning** and **Spatiotemporal Masking**. Brain Gradient Positioning introduces a functional coordinate system for brain functional parcellation, enhancing the positional encoding of different Regions of Interest (ROIs). Spatiotemporal Masking, tailored to the unique characteristics of fMRI data, addresses the challenge of heterogeneous time-series patches. These methodologies enhance model performance and advance our understanding of the neural circuits underlying cognition. Overall, Brain-JEPA is paving the way to address pivotal questions of building brain functional coordinate system and masking brain activity at the AI-neuroscience interface, and setting a potentially new paradigm in brain activity analysis through downstream adaptation.

Methods like PET imaging which use radiotracers that bind to amyloid-β (Aβ) - plaques that accumulate and lead to Alzheimer's disease (AD) - are crucial to detect AD early. However, current semi-quantitative methods that quantify Aβ using the Standardized Uptake Values Ratio (SUVr) cannot distinguish between the specific binding of radiotracers to Aβ and non-specific binding to other targets. In this paper, we propose a novel method to predict non-specific binding from MR images using deep learning by developing a novel conditional Generative Adversarial Network (cGAN) to improve Aβ quantification. PET images with low amyloid load were selected to represent non-specific uptake of the Aβ [11C]PiB radiotracer. Paired MR and PET images with low amyloid load were used for training, with the MRI image as the network input and the PET image as the training label. The cGAN was trained to learn the mapping between the MR image and the non-specific PET image. The generated non-specific PET images were compared to the real PET scans, achieving a mean SUVr difference of 1.90% in the gray matter. The small SUVr difference indicated that we have successfully developed cGAN for predicting the non-specific uptake of PET and that deep learning can be used to estimate the non-specific binding of Aβ PET radiotracer from MR images.

Quantification of cerebral white matter hyperintensities (WMH) of presumed vascular origin is of key importance in many neurological research studies. Currently, measurements are often still obtained from manual segmentations on brain MR images, which is a laborious procedure. The automatic WMH segmentation methods exist, but a standardized comparison of the performance of such methods is lacking. We organized a scientific challenge, in which developers could evaluate their methods on a standardized multi-center/-scanner image dataset, giving an objective comparison: the WMH Segmentation Challenge. Sixty T1 + FLAIR images from three MR scanners were released with the manual WMH segmentations for training. A test set of 110 images from five MR scanners was used for evaluation. The segmentation methods had to be containerized and submitted to the challenge organizers. Five evaluation metrics were used to rank the methods: 1) Dice similarity coefficient; 2) modified Hausdorff distance (95th percentile); 3) absolute log-transformed volume difference; 4) sensitivity for detecting individual lesions; and 5) F1-score for individual lesions. In addition, the methods were ranked on their inter-scanner robustness; 20 participants submitted their methods for evaluation. This paper provides a detailed analysis of the results. In brief, there is a cluster of four methods that rank significantly better than the other methods, with one clear winner. The inter-scanner robustness ranking shows that not all the methods generalize to unseen scanners. The challenge remains open for future submissions and provides a public platform for method evaluation.