Graph- and transformer-guided boundary aware network for medical image segmentation.

BACKGROUND AND OBJECTIVE: Despite the considerable progress achieved by U-Net-based models, medical image segmentation remains a challenging task due to complex backgrounds, irrelevant noises, and ambiguous boundaries. In this study, we present a novel approach called U-shaped Graph- and Transformer-guided Boundary Aware Network (GTBA-Net) to tackle these challenges. METHODS: GTBA-Net uses the pre-trained ResNet34 as its basic structure, and involves Global Feature Aggregation (GFA) modules for target localization, Graph-based Dynamic Feature Fusion (GDFF) modules for effective noise suppression, and Uncertainty-based Boundary Refinement (UBR) modules for accurate delineation of ambiguous boundaries. The GFA modules employ an efficient self-attention mechanism to facilitate coarse target localization amidst complex backgrounds, without introducing additional computational complexity. The GDFF modules leverage graph attention mechanism to aggregate information hidden among high- and low-level features, effectively suppressing target-irrelevant noises while preserving valuable spatial details. The UBR modules introduce an uncertainty quantification strategy and auxiliary loss to guide the model's focus towards target regions and uncertain "ridges", gradually mitigating boundary uncertainty and ultimately achieving accurate boundary delineation. RESULTS: Comparative experiments on five datasets encompassing diverse modalities (including X-ray, CT, endoscopic procedures, and ultrasound) demonstrate that the proposed GTBA-Net outperforms existing methods in various challenging scenarios. Subsequent ablation studies further demonstrate the efficacy of the GFA, GDFF, and UBR modules in target localization, noise suppression, and ambiguous boundary delineation, respectively. CONCLUSIONS: GTBA-Net exhibits substantial potential for extensive application in the field of medical image segmentation, particularly in scenarios involving complex backgrounds, target-irrelevant noises, or ambiguous boundaries.

Mitochondrial DNA induces nucleus pulposus cell pyroptosis via the TLR9-NF-κB-NLRP3 axis.

BACKGROUND: Nucleus pulposus cell (NPC) death and progressive reduction play important roles in intervertebral disc degeneration (IVDD). As part of a damage-associated molecular pattern, mitochondrial DNA (mtDNA) can be recognized by TLR9 and triggers the expression of NF-κB and NLRP3 inflammasomes, inducing pyroptosis and inflammatory response. However, whether mtDNA induces NPC pyroptosis via the TLR9-NF-κB-NLRP3 axis and promotes IVDD remains uncertain. METHODS: We constructed an in vitro NPC oxidative stress injury model to clarify the mechanism of mtDNA release, TLR9-NF-κB signaling pathway activation, and NPC injury. We further verified the mechanism of action underlying the inhibition of mtDNA release or TLR9 activation in NPC injury in vitro. We then constructed a rat punctured IVDD model to understand the mechanism inhibiting mtDNA release and TLR9 activation in IVDD. RESULTS: We used human NP specimen assays to show that the expression levels of TLR9, NF-κB, and NLRP3 inflammasomes correlated with the degree of IVDD. We demonstrated that mtDNA mediated TLR9-NF-κB-NLRP3 axis activation in oxidative stress-induced human NPC pyroptosis in vitro. Oxidative stress can damage the mitochondria of NPCs, causing the opening of the mitochondrial permeability transition pores (mPTP) and leading to the release of mtDNA into the cytosol. Furthermore, inhibition of mPTP opening or TLR9 activation blocked TLR9-NF-κB-NLRP3 axis activation and thereby mediated NPC pyroptosis and IVDD. CONCLUSION: mtDNA plays a key role in mediating NPC pyroptosis and IVDD via the TLR9-NF-κB-NLRP3 axis. Our findings provide new potential targets for IVDD.