img
img
Federated Learning for Privacy Preserving Abnormal Tooth Detection    
Yazarlar (3)
Ülkü Tuncer Küçüktaş
Dr. Öğr. Üyesi Fatih UYSAL Dr. Öğr. Üyesi Fatih UYSAL
Kafkas Üniversitesi, Türkiye
Fırat Hardalaç
Gazi Üniversitesi, Türkiye
Devamını Göster
Özet
Federated learning has become a key method in medical research, not only for its effectiveness in handling sensitive data but also for enabling collaboration between different centers while ensuring patient privacy. This study aims to fill a research gap in the use of federated learning in dentistry, a field where the protection of sensitive data is crucial. We compare the performance of federated learning with local and central learning in dental disease detection. Our research focuses on two scenarios: one involving imbalanced data distribution across different centers, and another on continual learning, where the model's ability to learn new classes introduced by a single center is evaluated. Our results indicate that federated learning could be a promising approach, particularly when central learning is not feasible. The findings are especially encouraging in the continual learning scenario, suggesting that federated learning is adept at adapting to new information while maintaining previously acquired knowledge.
Anahtar Kelimeler
carries | continual learning | deep learning | federated learning | periapical lesion
Bildiri Türü Tebliğ/Bildiri
Bildiri Alt Türü Tam Metin Olarak Yayımlanan Tebliğ (Uluslararası Kongre/Sempozyum)
Bildiri Niteliği Alanında Hakemli Uluslararası Kongre/Sempozyum
Bildiri Dili İngilizce
Kongre Adı 11th International Conference on Electrical and Electronics Engineering (ICEEE)
Kongre Tarihi 22-04-2024 / 24-04-2024
Basıldığı Ülke Türkiye
Basıldığı Şehir Muğla
BM Sürdürülebilir Kalkınma Amaçları
Atıf Sayıları
Federated Learning for Privacy Preserving Abnormal Tooth Detection

Paylaş