Diffusion MRI data analysis assisted by deep learning synthesized anatomical images (DeepAnat)

• Published in: Medical image analysis 2023-05, Vol.86, p.102744-102744, Article 102744
• Main Authors: Li, Ziyu, Fan, Qiuyun, Bilgic, Berkin, Wang, Guangzhi, Wu, Wenchuan, Polimeni, Jonathan R., Miller, Karla L., Huang, Susie Y., Tian, Qiyuan
• Format: Article
• Language: English
• Subjects: Aged; Brain segmentation; Convolutional neural network; Cortical surface reconstruction; Data Analysis; Deep Learning; Diffusion Magnetic Resonance Imaging - methods; Diffusion tractography; Generative adversarial network; Humans; Image co-registration; Image Processing, Computer-Assisted - methods; Magnetic Resonance Imaging - methods
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2023.102744

•T1w images synthesized from diffusion data using CNNs exhibit high quality.•Resultant volumetric segmentation and cortical surfaces are accurate for various diffusion analyses.•U-Net produces higher segmentation accuracy than GAN.•U-Net is generalizable to datasets with different spatial resolution and b-value and can be improved by fine-tuning.•Synthesized T1w images improve co-registration between diffusion images with geometric distortion and T1w data. Diffusion MRI is a useful neuroimaging tool for non-invasive mapping of human brain microstructure and structural connections. The analysis of diffusion MRI data often requires brain segmentation, including volumetric segmentation and cerebral cortical surfaces, from additional high-resolution T1-weighted (T1w) anatomical MRI data, which may be unacquired, corrupted by subject motion or hardware failure, or cannot be accurately co-registered to the diffusion data that are not corrected for susceptibility-induced geometric distortion. To address these challenges, this study proposes to synthesize high-quality T1w anatomical images directly from diffusion data using convolutional neural networks (CNNs) (entitled “DeepAnat”), including a U-Net and a hybrid generative adversarial network (GAN), and perform brain segmentation on synthesized T1w images or assist the co-registration using synthesized T1w images. The quantitative and systematic evaluations using data of 60 young subjects provided by the Human Connectome Project (HCP) show that the synthesized T1w images and results for brain segmentation and comprehensive diffusion analysis tasks are highly similar to those from native T1w data. The brain segmentation accuracy is slightly higher for the U-Net than the GAN. The efficacy of DeepAnat is further validated on a larger dataset of 300 more elderly subjects provided by the UK Biobank. Moreover, the U-Nets trained and validated on the HCP and UK Biobank data are shown to be highly generalizable to the diffusion data from Massachusetts General Hospital Connectome Diffusion Microstructure Dataset (MGH CDMD) acquired with different hardware systems and imaging protocols and therefore can be used directly without retraining or with fine-tuning for further improved performance. Finally, it is quantitatively demonstrated that the alignment between native T1w images and diffusion images uncorrected for geometric distortion assisted by synthesized T1w images substantially improves upon that by directly co-regist