TY - GEN
T1 - Cluster Based Secure Multi-party Computation in Federated Learning for Histopathology Images
AU - Hosseini, Seyedeh Maryam
AU - Sikaroudi, Milad
AU - Babaei, Morteza
AU - Tizhoosh, Hamid R.
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - Federated learning (FL) is a decentralized method enabling hospitals to collaboratively learn a model without sharing private patient data for training. In FL, participant hospitals periodically exchange training results rather than training samples with a central server. However, having access to model parameters or gradients can expose private training data samples. To address this challenge, we adopt secure multiparty computation (SMC) to establish a privacy-preserving federated learning framework. In our proposed method, the hospitals are divided into clusters. After local training, each hospital splits its model weights among other hospitals in the same cluster such that no single hospital can retrieve other hospitals’ weights on its own. Then, all hospitals sum up the received weights, sending the results to the central server. Finally, the central server aggregates the results, retrieving the average of models’ weights and updating the model without having access to individual hospitals’ weights. We conduct experiments on a publicly available repository, The Cancer Genome Atlas (TCGA). We compare the performance of the proposed framework with differential privacy and federated averaging as the baseline. The results reveal that compared to differential privacy, our framework can achieve higher accuracy with no privacy leakage risk at a cost of higher communication overhead.
AB - Federated learning (FL) is a decentralized method enabling hospitals to collaboratively learn a model without sharing private patient data for training. In FL, participant hospitals periodically exchange training results rather than training samples with a central server. However, having access to model parameters or gradients can expose private training data samples. To address this challenge, we adopt secure multiparty computation (SMC) to establish a privacy-preserving federated learning framework. In our proposed method, the hospitals are divided into clusters. After local training, each hospital splits its model weights among other hospitals in the same cluster such that no single hospital can retrieve other hospitals’ weights on its own. Then, all hospitals sum up the received weights, sending the results to the central server. Finally, the central server aggregates the results, retrieving the average of models’ weights and updating the model without having access to individual hospitals’ weights. We conduct experiments on a publicly available repository, The Cancer Genome Atlas (TCGA). We compare the performance of the proposed framework with differential privacy and federated averaging as the baseline. The results reveal that compared to differential privacy, our framework can achieve higher accuracy with no privacy leakage risk at a cost of higher communication overhead.
KW - Decentralized learning
KW - Federated learning
KW - Histopathology imaging
KW - Privacy preservation
KW - Secure multiparty computation
UR - http://www.scopus.com/inward/record.url?scp=85141765220&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85141765220&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-18523-6_11
DO - 10.1007/978-3-031-18523-6_11
M3 - Conference contribution
AN - SCOPUS:85141765220
SN - 9783031185229
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 110
EP - 118
BT - Distributed, Collaborative, and Federated Learning, and Affordable AI and Healthcare for Resource Diverse Global Health - 3rd MICCAI Workshop, DeCaF 2022, and 2nd MICCAI Workshop, FAIR 2022, Held in Conjunction with MICCAI 2022, Proceedings
A2 - Albarqouni, Shadi
A2 - Bakas, Spyridon
A2 - Bano, Sophia
A2 - Cardoso, M. Jorge
A2 - Khanal, Bishesh
A2 - Landman, Bennett
A2 - Li, Xiaoxiao
A2 - Qin, Chen
A2 - Rekik, Islem
A2 - Rieke, Nicola
A2 - Roth, Holger
A2 - Xu, Daguang
A2 - Sheet, Debdoot
PB - Springer Science and Business Media Deutschland GmbH
T2 - 3rd MICCAI Workshop on Distributed, Collaborative, and Federated Learning, DeCaF 2022, and the 2nd MICCAI Workshop on Affordable AI and Healthcare, FAIR 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022
Y2 - 22 September 2022 through 22 September 2022
ER -