Active trachoma: enhancing image classification using pretrained SOTA models and explainable AI
Frontiers in Bacteriology · 2024
DOI: 10.3389/fbrio.2024.1333641 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Bacteriology, 2024. Verfasst von Yongjun Pan und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Trachoma, an infectious disease that leads to blindness, continues to pose a significant public health challenge in over 40 countries as of 2023. The initial phase of this disease, “active trachoma” is characterized by inflammation and can be effectively treated with non-surgical interventions. However, if left untreated, it progresses to the “scarring” phase, often requiring surgical intervention. Earlier detection of “active trachoma” is critical to prevent unnecessary surgery and also to reduce the transmission of the infection. Developing accessible tools for a region with limited resources is necessary. Deep neural networks have proven their effectiveness in numerous image and vision-related tasks, yet research on “active trachoma” has received still little attention.
Method
In this study, we adapted several pre-trained state-of-the-art deep neural network models like ResNet, Xception from image classification on “active classification” task. Further experiments were also conducted in three cases: training from scratch, training from pretrained models on raw images and on region-of-interest (ROI) focused images. Results and discussion The results indicate that these models outperformed the previous studies using the same dataset, achieving an improvement of 6\% on detection of follicular trachomatous inflammation and 12\% for detection of intense trachomatous inflammation. Furthermore, we employed the eXplainable Artificial Intelligence tool Grad-CAM, which revealed a significant discrepancy between eyelid's geometric centroid and attention centroid from models with high classification accuracy. This finding suggests that the conventional method of selecting a region of interest based on the geometric centroid may need to be adjusted. Using XAI can offer valuable insights into understanding the classification and progression of active trachoma.
Zitieren
APA
Pan, Y., Lan, W., & Xu, B. (2024). Active trachoma: enhancing image classification using pretrained SOTA models and explainable AI. Frontiers in Bacteriology. https://doi.org/10.3389/fbrio.2024.1333641
Vancouver
Pan Y, Lan W, Xu B. Active trachoma: enhancing image classification using pretrained SOTA models and explainable AI. Frontiers in Bacteriology. 2024. doi:10.3389/fbrio.2024.1333641
BibTeX
@article{Pan2024,
author = {Pan, Yongjun and Lan, Wenyao and Xu, Binbin},
title = {Active trachoma: enhancing image classification using pretrained SOTA models and explainable AI},
journal = {Frontiers in Bacteriology},
year = {2024},
doi = {10.3389/fbrio.2024.1333641}
}
EUVIMED