Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study presents AppendiGrade, an advanced deep learning framework that utilizes ultrasound images to accurately classify complicated appendicitis cases, addressing a critical challenge in emergency medicine. By employing four pretrained models and optimizing them through techniques such as hyperparameter tuning and image preprocessing, the InceptionV3 model achieved a remarkable accuracy of 95.58% in detecting complicated appendicitis. Additionally, the integration of Grad-CAM enhances interpretability by visualizing the model's focus areas, facilitating expert validation of the predictions.
Achieving 95.58% accuracy in detecting complicated appendicitis from ultrasound images could revolutionize emergency diagnostics and improve patient outcomes.
Appendicitis is one of the most common abdominal emergencies worldwide and requires prompt diagnosis and treatment to prevent life-threatening conditions. However, accurately differentiating complicated cases, such as perforation or abscess formation, from uncomplicated appendicitis remains a significant clinical challenge. Among other methods, ultrasound is a safer and more cost-efficient diagnostic technique because of the lack of radiation exposure. In this research, an advanced system capable of automatically detecting complicated appendicitis from ultrasound images was developed. A dataset consisting of 4679 ultrasound images with 5 classes, namely perforated, abscess, acute, appendicolith, and normal, was used for the proposed model training and testing. Four pretrained deep learning models, DenseNet201, InceptionV3, ConvNextTiny, and VGG19, have been employed for detecting and classifying complicated appendicitis. In the initial configuration, InceptionV3 achieved the second highest accuracy, with a value of 69.21%. Owing to suboptimal performance with raw images, further optimization techniques, including image preprocessing, hyperparameter tuning, model fine-tuning, and image sharpening, were applied. These enhancements significantly improved the model's performance, with an accuracy of 95.58% for InceptionV3. The model performance is then explained with gradient-weighted class activation mapping (Grad-CAM), which creates a heatmap of the regions responsible for the model's prediction of the infected areas. This could make crosschecking with experts much easier.