RESUMO
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.
RESUMO
With the global emergence of drug-resistant bacteria causing difficult-to-treat infections, there is an urgent need for a tool to facilitate studies on key virulence and antimicrobial resistant factors. Mass spectrometry (MS) has contributed substantially to the elucidation of the structure-function relationships of lipid A, the endotoxic component of lipopolysaccharide which also serves as an important protective barrier against antimicrobials. Here, we present LipidA-IDER, an automated structure annotation tool for system-level scale identification of lipid A from high-resolution tandem mass spectrometry (MS2) data. LipidA-IDER was validated against previously reported structures of lipid A in the reference bacteria, Escherichia coli and Pseudomonas aeruginosa. Using MS2 data of variable quality, we demonstrated LipidA-IDER annotated lipid A with a performance of 71.2% specificity and 70.9% sensitivity, offering greater accuracy than existing lipidomics software. The organism-independent workflow was further applied to a panel of six bacterial species: E. coli and Gram-negative members of ESKAPE pathogens. A comprehensive atlas comprising 188 distinct lipid A species, including remodeling intermediates, was generated and can be integrated with software including MS-DIAL and Metabokit for identification and semiquantitation. Systematic comparison of a pair of polymyxin-sensitive and polymyxin-resistant Acinetobacter baumannii isolated from a human patient unraveled multiple key lipid A structural features of polymyxin resistance within a single analysis. Probing the lipid A landscape of bacteria using LipidA-IDER thus holds immense potential for advancing our understanding of the vast diversity and structural complexity of a key lipid virulence and antimicrobial-resistant factor. LipidA-IDER is freely available at https://github.com/Systems-Biology-Of-Lipid-Metabolism-Lab/LipidA-IDER.