Contrast-Informed Augmentation and Domain-Adversarial Training Improve MRI Reconstruction for Neonatal Imaging
Researchers developed and tested training strategies to improve deep learning-based MRI reconstruction models trained on adult brain scans so they generalize better to neonatal brain imaging. The study compared three training regimes—adult-only, mixed augmentation, and mixed domain-adversarial training—on the E2E-VarNet architecture using retrospectively undersampled T2-weighted MRI data. The findings suggest that combining contrast-informed data augmentation with adversarial training can meaningfully reduce the performance gap caused by domain shift between adult and neonatal MRI.
A preprint posted to arXiv investigates whether contrast-informed data augmentation and domain-adversarial training can help MRI reconstruction models trained on adult brain data perform better when applied to neonatal brain scans, a clinically important but data-scarce population. Three training regimes were compared: adult-only training with unaugmented data, mixed training incorporating neonatal-informed augmented adult data, and mixed training with an added domain-adversarial objective. Models were evaluated at acceleration factors R=4 and R=8 using SSIM and PSNR metrics on both neonatal and adult test sets. At R=4, the mixed domain-adversarial training (Mixed-DAT) approach achieved the best neonatal performance with an SSIM of 0.924 and PSNR of 33.98 dB, compared to 0.766 SSIM for the unaugmented adult-only baseline at R=8. Feature-space analysis using t-SNE plots indicated that domain-adversarial training brought the latent representations of adult, augmented adult, and neonatal samples closer together, suggesting improved domain alignment. The authors conclude that these techniques offer a promising path toward more robust undersampled neonatal MRI reconstruction without requiring large neonatal training datasets.
What's missing
The study uses retrospectively undersampled data rather than prospectively acquired accelerated scans, which may not fully reflect real-world scanner acquisition conditions. The neonatal dataset size and demographic diversity are not described in the abstract, limiting assessment of generalizability. It is also unclear whether the augmentation strategy was validated across MRI field strengths or scanner vendors beyond those used in the study.
What different sources said
- arXiv cs.AICenter
Contrast-Informed Augmentation and Domain-Adversarial Training for Adult-to-Neonatal MR Reconstruction Generalization
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.