Search

Large data initiatives and high-powered brain imaging analyses require the pooling of MR images acquired across multiple scanners, often using different protocols. Prospective cross-site harmonization often involves the use of a phantom or traveling subjects. However, as more datasets are becoming publicly available, there is a growing need for retrospective harmonization, pooling data from sites not originally coordinated together. Several retrospective harmonization techniques have shown promise in removing cross-site image variation. However, most unsupervised methods cannot distinguish between image-acquisition based variability and cross-site population variability, so they require that datasets contain subjects or patient groups with similar clinical or demographic information. To overcome this limitation, we consider cross-site MRI image harmonization as a style transfer problem rather than a domain transfer problem. Using a fully unsupervised deep-learning framework based on a generative adversarial network (GAN), we show that MR images can be harmonized by inserting the style information encoded from a reference image directly, without knowing their site/scanner labelsa priori. We trained our model using data from five large-scale multi-site datasets with varied demographics. Results demonstrated that our styleencoding model can harmonize MR images, and match intensity profiles, successfully, without relying on traveling subjects. This model also avoids the need to control for clinical, diagnostic, or demographic information. Moreover, we further demonstrated that if we included diverse enough images into the training set, our method successfully harmonized MR images collected from unseen scanners and protocols, suggesting a promising novel tool for ongoing collaborative studies.

Diffusion-weighted magnetic resonance imaging (dMRI) is sensitive to microstructural changes in the brain that occur with normal aging and Alzheimer’s disease (AD). There is much interest in which dMRI measures are most strongly correlated with clinical measures of AD severity, such as the clinical dementia rating (CDR), and biological processes that may be disrupted in AD, such as brain amyloid load measured using PET. Of these processes, some can be targeted using novel drugs. Since 2016, the Alzheimer’s Disease Neuroimaging Initiative (ADNI) has collected dMRI data from three scanner manufacturers across 58 sites using 7 different protocols that vary in angular resolution, scan duration, and in the number and distribution of diffusion-weighted gradients. Here, we assessed dMRI data from 730 of those individuals (447 cognitively normal controls, 214 with mild cognitive impairment, 69 with dementia; age: 74.1±7.9 years; 381 female/349 male). To harmonize data from different protocols, we applied ComBat, ComBat-GAM, and CovBat to dMRI metrics from 28 white matter regions of interest. We ranked all dMRI metrics in order of the strength of clinically relevant associations, and assessed how this depended on the harmonization methods employed. dMRI metrics were associated with age and clinical impairment, but also with amyloid positivity. All harmonization methods gave comparable results while enabling data integration across multiple scanners and protocols.

Quality control (QC) is a vital step for all scientific data analyses and is critically important in the biomedical sciences. Image segmentation is a common task in medical image analysis, and automated tools to segment many regions from human brain MRIs are now well established. However, these methods do not always give anatomically correct labels. Traditional methods for QC tend to reject statistical outliers, which may not necessarily be inaccurate. Here, we make use of a large database of over 12,000 brain images that contain 68 parcellations of the human cortex, each of which was assessed for anatomical accuracy by a human rater. We trained three machine learning models to determine if a region was anatomically accurate (as ‘pass’, or ‘fail’) and tested the performance on an independent dataset. We found good performance for the majority of labeled regions. This work will facilitate more anatomically accurate large-scale multi-site research.