C2PSA-Enhanced YOLOv11: Boosting Pneumonia Detection in Chest X-rays Through Pseudo-Color Attention Mechanisms
Year: 2025
Type: article
Abstract: Pneumonia, a common respiratory infection, has a serious impact on human lives worldwide. In this paper, the editors propose a Yolov11 deep learning model that introduces an attention mechanism and present a data enhancement technique for medical imaging. After training and tuning on a public dataset and doing comparative experiments with other deep learning models, the model can efficiently detect lung X-images and determine whether it has pneumonia or not. The experimental data show that the accuracy, precision, recall, and F1-score of the model are 90.12%, 91.45%, 90.66%, and 91.05%, respectively, and have excellent results in the comparison experiments with four common architectures, namely, Yolov11, Yolov8, ResNet18, and ResNet50, which confirms the superiority of the model. By comparing the performance of the model before and after data enhancement, it was concluded that pseudo-coloring of gray medical images can improve the accuracy of the model by 4-7%.
Source: Biomedical Data Science
Author: Yuchen Li
Institution: Guangdong Polytechnic Normal University
Language: English
Topic: COVID-19 diagnosis using AI
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4413026617