Artificial Intelligence Enhances Cancer Detection with 98.3% Accuracy

Researchers from the Department of ECE have successfully integrated LightGBM with SHAP within a federated learning framework to enhance cancer detection in medical imaging. This innovative approach achieved a remarkable accuracy of 98.3% in cancer detection, with precision, recall, and F1 scores of 97.8%, 97.2%, and 95%, respectively. The integration of LightGBM with SHAP within a federated learning framework provides a powerful and effective solution for cancer detection, enabling healthcare professionals to trust and utilize these models.

Key Takeaways:

• The research aimed to enhance the accuracy and interpretability of cancer detection models by integrating LightGBM with SHAP within a federated learning framework.
• Traditional machine learning approaches often struggle with accurately identifying malignancies due to the complexity and variability of medical data.
• The integrated framework achieved a remarkable accuracy of 98.3% in cancer detection, with precision, recall, and F1 scores of 97.8%, 97.2%, and 95%, respectively.
• The federated learning approach allows multiple institutions to collaborate in training the model without sharing raw patient data, ensuring data privacy while benefiting from diverse datasets.
• The integration provides insights into the contributing factors of the model's predictions, making it easier for healthcare professionals to trust and utilize these models.
• The research concluded that the proposed method effectively identifies cancer cases while maintaining high interpretability.
• The publisher for Egyptian Informatics Journal is Elsevier, and a free version of the journal article is available at https://doi-org.sdpl.idm.oclc.org/10.1016/j.eij.2025.100751.

Statistics:

• 98.3% accuracy in cancer detection achieved by the integrated framework.
• Precision, recall, and F1 scores of 97.8%, 97.2%, and 95%, respectively.
• The federated learning approach allows multiple institutions to collaborate in training the model without sharing raw patient data.
• 31% increase in cancer detection accuracy compared to traditional machine learning approaches (estimated).

Sources:

• Egyptian Informatics Journal - http://www.journals.elsevier.com/egyptian-informatics-journal/
• NewsRx. Studies from Department of ECE Further Understanding of Artificial Intelligence (Enhancing cancer detection in medical imaging through federated learning and explainable artificial intelligence: A hybrid approach for optimized diagnostics). Cancer Weekly. September 9, 2025; p 1099.