Self-Supervised Vision Transformers Achieve High Accuracy in Detecting Temporomandibular Joint Osteoarthritis on CBCT Scans
Researchers developed a pipeline using DINO-family self-supervised vision transformers to automatically detect temporomandibular joint osteoarthritis (TMJ OA) from cone-beam CT scans, achieving an AUC of 0.902. The study found that partially unfreezing the final two transformer blocks of a pretrained model was the key factor driving performance gains, outperforming fully frozen, fully fine-tuned, and supervised baseline approaches. The findings offer practical guidance for deploying large vision foundation models in medical imaging contexts where labeled data is scarce.
A preprint posted to arXiv presents a slice-based deep learning pipeline for detecting TMJ osteoarthritis from cone-beam CT (CBCT) images, leveraging the DINO family of self-supervised vision transformers. The system encodes individual axial CBCT slices with a Vision Transformer backbone and aggregates predictions across slices using attention-based multiple instance learning for patient-level binary classification. Through systematic ablation experiments, the authors found that partially unfreezing only the final two transformer blocks of DINOv2 raised the area under the ROC curve from 0.671 (fully frozen) to 0.902, surpassing DINOv1 (0.867), DINOv2 with registers (0.774), and a supervised ImageNet-pretrained ViT-B/16 baseline (0.843). Notably, RAD-DINO, a radiology-domain pretrained variant, did not outperform general-purpose DINOv2 with partial unfreezing, suggesting that adaptation strategy matters more than domain-specific pretraining. The study was conducted on a multi-source CBCT dataset, and the authors argue their results provide actionable guidance for adapting foundation models to low-data medical imaging tasks.
What's missing
The paper does not report the size or demographic composition of the multi-source CBCT dataset, making it difficult to assess generalizability. The study does not include external validation on an independent held-out cohort, and clinical specificity, sensitivity, and calibration metrics beyond AUC are not prominently reported. As a preprint, the work has not yet undergone peer review. Open questions include whether the approach generalizes across different CBCT scanner manufacturers and acquisition protocols, and how performance compares to trained radiologists or oral and maxillofacial specialists.
What different sources said
- arXiv cs.AICenter
Self-Supervised Vision Transformers for CBCT-Based Detection of Temporomandibular Joint Osteoarthritis
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.