A Deep Learning-Driven Inhalation Injury Grading Assistant Using Bronchoscopy Images

TL;DR

This paper presents a deep learning-based tool for grading inhalation injuries from bronchoscopy images, improving accuracy and consistency over traditional subjective methods by using advanced data augmentation and model analysis techniques.

Contribution

Introduces a novel deep learning diagnostic assistant utilizing data augmentation and model evaluation to improve inhalation injury grading accuracy from bronchoscopy images.

Findings

Achieved 97.8% classification accuracy with GoogLeNet and CUT.

Demonstrated that data augmentation improves class separability and model performance.

Provided insights into model decision-making via Grad-CAM analysis.

Abstract

Inhalation injuries present a challenge in clinical diagnosis and grading due to Conventional grading methods such as the Abbreviated Injury Score (AIS) being subjective and lacking robust correlation with clinical parameters like mechanical ventilation duration and patient mortality. This study introduces a novel deep learning-based diagnosis assistant tool for grading inhalation injuries using bronchoscopy images to overcome subjective variability and enhance consistency in severity assessment. Our approach leverages data augmentation techniques, including graphic transformations, Contrastive Unpaired Translation (CUT), and CycleGAN, to address the scarcity of medical imaging data. We evaluate the classification performance of two deep learning models, GoogLeNet and Vision Transformer (ViT), across a dataset significantly expanded through these augmentation methods. The results demonstrate GoogLeNet combined with CUT as the most effective configuration for grading inhalation injuries through bronchoscopy images and achieves a classification accuracy of 97.8%. The histograms and frequency analysis evaluations reveal variations caused by the augmentation CUT with distribution changes in the histogram and texture details of the frequency spectrum. PCA visualizations underscore the CUT substantially enhances class separability in the feature space. Moreover, Grad-CAM analyses provide insight into the decision-making process; mean intensity for CUT heatmaps is 119.6, which significantly exceeds 98.8 of the original datasets. Our proposed tool leverages mechanical ventilation periods as a novel grading standard, providing comprehensive diagnostic support.