Abstract Listings 2025

Purpose

To develop a deep learning-based system for automatically identifying and classifying benign and malignant tumors of the eyelid in order to improve diagnostic accuracy and efficiency.

Methods

The dataset includes photographs of normal eyelids, benign and malignant eyelid tumors and was randomly divided into a training and validation dataset in a ratio of 8:2. We used the training dataset to train eight convolutional neural network (CNN) models to classify normal eyelids, benign and malignant eyelid tumors. These models included VGG16, ResNet50, Inception-v4, EfficientNet-V2-M and their variants. The validation dataset was utilized to evaluate and compare the performance of the different deep learning models.

Results

All eight models achieved an average accuracy greater than 0.746 for identifying normal eyelids, benign and malignant eyelid tumors, with an average sensitivity and specificity exceeding 0.790 and 0.866, respectively. The mean area under the receiver operating characteristic curve (AUC) for the eight models was more than 0.904 in correctly identifying normal eyelids, benign and malignant eyelid tumors. The dual-path Inception-v4 network demonstrated the highest performance, with an AUC of 0.930 (95%CI 0.900-0.954) and an F1-score of 0.838 (95%CI 0.787-0.882).

Conclusion

The deep learning-based system shows significant potential in improving the diagnosis of eyelid tumors, providing a reliable and efficient tool for clinical practice. Future work will validate the model with more extensive and diverse datasets and integrate it into clinical workflows for real-time diagnostic support.

Additional Authors

↑ Back to top