Rabies diagnosis in low-data settings: A comparative study on the impact of data augmentation and transfer learning

arXiv cs.CV / 4/23/2026

📰 NewsDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

共有:

Key Points

The study addresses the difficulty of rabies diagnosis in low-data, resource-limited settings where fluorescence microscopy requires scarce expert interpretation.
It proposes an AI-driven automated diagnostic pipeline using transfer learning with four deep learning architectures (EfficientNetB0, EfficientNetB2, VGG16, and ViT-B16) and fluorescent image analysis.
Three data augmentation strategies were compared, with TrivialAugmentWide found to best preserve key fluorescent patterns while improving robustness and generalization.
On a small dataset of 155 microscopic images, the EfficientNetB0 model with specific geometric and color augmentation (selected via stratified 3-fold cross-validation) delivered the best classification performance on cropped images despite class imbalance.
The authors deployed an online tool to support practical use and position the approach for further optimization toward broader medical imaging applications.

Abstract

Rabies remains a major public health concern across many African and Asian countries, where accurate diagnosis is critical for effective epidemiological surveillance. The gold standard diagnostic methods rely heavily on fluorescence microscopy, necessitating skilled laboratory personnel for the accurate interpretation of results. Such expertise is often scarce, particularly in regions with low annual sample volumes. This paper presents an automated, AI-driven diagnostic system designed to address these challenges. We developed a robust pipeline utilizing fluorescent image analysis through transfer learning with four deep learning architectures: EfficientNetB0, EfficientNetB2, VGG16, and Vision Transformer (ViTB16). Three distinct data augmentation strategies were evaluated to enhance model generalization on a dataset of 155 microscopic images (123 positive and 32 negative). Our results demonstrate that TrivialAugmentWide was the most effective augmentation technique, as it preserved critical fluorescent patterns while improving model robustness. The EfficientNetB0 model, utilizing Geometric & Color augmentation and selected through stratified 3fold cross-validation, achieved optimal classification performance on cropped images. Despite constraints posed by class imbalance and a limited dataset size, this work confirms the viability of deep learning for automating rabies diagnosis. The proposed method enables fast and reliable detection with significant potential for further optimization. An online tool was deployed to facilitate practical access, establishing a framework for future medical imaging applications. This research underscores the potential of optimized deep learning models to transform rabies diagnostics and improve public health outcomes.