Comparative Analysis of Vision Transformers and Conventional Convolutional Neural Networks in Detecting Referable Diabetic Retinopathy.

Objective: Vision transformers (ViTs) have shown promising performance in various classification tasks previously dominated by convolutional neural networks (CNNs). However, the performance of ViTs in referable diabetic retinopathy (DR) detection is relatively underexplored. In this study, using retinal photographs, we evaluated the comparative performances of ViTs and CNNs on detection of referable DR. Design: Retrospective study. Participants: A total of 48 269 retinal images from the open-source Kaggle DR detection dataset, the Messidor-1 dataset and the Singapore Epidemiology of Eye Diseases (SEED) study were included. Methods: Using 41 614 retinal photographs from the Kaggle dataset, we developed 5 CNN (Visual Geometry Group 19, ResNet50, InceptionV3, DenseNet201, and EfficientNetV2S) and 4 ViTs models (VAN_small, CrossViT_small, ViT_small, and Hierarchical Vision transformer using Shifted Windows [SWIN]_tiny) for the detection of referable DR. We defined the presence of referable DR as eyes with moderate or worse DR. The comparative performance of all 9 models was evaluated in the Kaggle internal test dataset (with 1045 study eyes), and in 2 external test sets, the SEED study (5455 study eyes) and the Messidor-1 (1200 study eyes). Main Outcome Measures: Area under operating characteristics curve (AUC), specificity, and sensitivity. Results: Among all models, the SWIN transformer displayed the highest AUC of 95.7% on the internal test set, significantly outperforming the CNN models (all P < 0.001). The same observation was confirmed in the external test sets, with the SWIN transformer achieving AUC of 97.3% in SEED and 96.3% in Messidor-1. When specificity level was fixed at 80% for the internal test, the SWIN transformer achieved the highest sensitivity of 94.4%, significantly better than all the CNN models (sensitivity levels ranging between 76.3% and 83.8%; all P < 0.001). This trend was also consistently observed in both external test sets. Conclusions: Our findings demonstrate that ViTs provide superior performance over CNNs in detecting referable DR from retinal photographs. These results point to the potential of utilizing ViT models to improve and optimize retinal photo-based deep learning for referable DR detection. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Inflammasome signaling is dispensable for ß-amyloid-induced neuropathology in preclinical models of Alzheimer's disease.

Background: Alzheimer's disease (AD) is the most common neurodegenerative disorder affecting memory and cognition. The disease is accompanied by an abnormal deposition of ß-amyloid plaques in the brain that contributes to neurodegeneration and is known to induce glial inflammation. Studies in the APP/PS1 mouse model of ß-amyloid-induced neuropathology have suggested a role for inflammasome activation in ß-amyloid-induced neuroinflammation and neuropathology. Methods: Here, we evaluated the in vivo role of microglia-selective and full body inflammasome signalling in several mouse models of ß-amyloid-induced AD neuropathology. Results: Microglia-specific deletion of the inflammasome regulator A20 and inflammasome effector protease caspase-1 in the AppNL-G-F and APP/PS1 models failed to identify a prominent role for microglial inflammasome signalling in ß-amyloid-induced neuropathology. Moreover, global inflammasome inactivation through respectively full body deletion of caspases 1 and 11 in AppNL-G-F mice and Nlrp3 deletion in APP/PS1 mice also failed to modulate amyloid pathology and disease progression. In agreement, single-cell RNA sequencing did not reveal an important role for Nlrp3 signalling in driving microglial activation and the transition into disease-associated states, both during homeostasis and upon amyloid pathology. Conclusion: Collectively, these results question a generalizable role for inflammasome activation in preclinical amyloid-only models of neuroinflammation.

Inflammasomes in neuroinflammatory and neurodegenerative diseases.

Neuroinflammation and neurodegeneration often result from the aberrant deposition of aggregated host proteins, including amyloid-ß, α-synuclein, and prions, that can activate inflammasomes. Inflammasomes function as intracellular sensors of both microbial pathogens and foreign as well as host-derived danger signals. Upon activation, they induce an innate immune response by secreting the inflammatory cytokines interleukin (IL)-1ß and IL-18, and additionally by inducing pyroptosis, a lytic cell death mode that releases additional inflammatory mediators. Microglia are the prominent innate immune cells in the brain for inflammasome activation. However, additional CNS-resident cell types including astrocytes and neurons, as well as infiltrating myeloid cells from the periphery, express and activate inflammasomes. In this review, we will discuss current understanding of the role of inflammasomes in common degenerative diseases of the brain and highlight inflammasome-targeted strategies that may potentially treat these diseases.