Deep Learning‐Driven Transformation: A Novel Approach for Mitigating Batch Effects in Diffusion MRI Beyond Traditional Harmonization

Abstract

Background“Batch effect” in MR images, due to vendor‐specific features, MR machine generations, and imaging parameters, challenges image quality and hinders deep learning (DL) model generalizability.PurposeWe aim to develop a DL model using contrast adjustment and super‐resolution to reduce diffusion‐weighted images (DWIs) diversity across magnetic field strengths and imaging parameters.Study TypeRetrospective.SubjectsThe DL model was built using an open dataset from one individual. The MR machine identification model was trained and validated on a dataset of 1134 adults (54% females, 46% males), with 1050 subjects showing no DWI abnormalities and 84 with conditions like stroke and tumors. The 21,000 images were divided into 80% for training, 20% for validation, and 3500 for testing.Field Strength/SequenceSeven MR scanners from four manufacturers with 1.5 T and 3 T magnetic field strengths. DWIs were acquired using spin‐echo sequences and high‐resolution T2WIs using the T2‐SPACE sequence.AssessmentAn experienced, board‐certified radiologist evaluated the effectiveness of restoring high‐resolution T2WI and harmonizing diverse DWI with metrics such as PSNR and SSIM, and the texture and frequency attributes were further analyzed using gray‐level co‐occurrence matrix and 1‐dimensional power spectral density. The model's impact on machine‐specific characteristics was gauged through the performance metrics of a ResNet‐50 model. Comprehensive statistical tests were employed for statistical robustness, including McNemar's test and the Dice index.ResultsOur DL protocol reduced DWI contrast and resolution variation. ResNet‐50 model's accuracy decreased from 0.9443 to 0.5786, precision from 0.9442 to 0.6494, recall from 0.9443 to 0.5786, and F1 score from 0.9438 to 0.5587. The t‐SNE visualization indicated more consistent image features across multiple MR devices. Autoencoder halved learning iterations; Dice coefficient >0.74 confirmed signal reproducibility in 84 lesions.ConclusionThis study presents a DL strategy to mitigate batch effects in diffusion MR images, improving their quality and generalizability.Evidence Level3Technical EfficacyStage 1

Harmonizing multisite neonatal diffusion-weighted brain MRI data for developmental neuroscience

“…Future studies should investigate the effect of ComBat on multi-centre neonatal dMRI case-control studies. In addition, further research is needed to assess the utility of novel harmonization methods such as deep learning techniques (Wada et al, 2023) for neonatal dMRI. In this study we chose to focus on harmonization of DTI metrics as these are widely used in neonatal dMRI research (Pecheva et al, 2018), and can be calculated from many conventional clinical dMRI acquisitions.…”