Table 5 Results of communication and computational overhead of the proposed method over different datasets.

From: A privacy-preserving expert system for collaborative medical diagnosis across multiple institutions using federated learning

Metric

Dermatology32

HCV33

Diabetes Risk34

PHC Operations

(ci=Epk(Di) (ms)

150

200

220

FL-RDBN Computations

(Ci=FL-RDBN(M,ci)

1000

1100

1150

Data Transfer

(User to Server)

(ci) (MB)

5

6

6.5

Data Transfer

(Server to User)

(cθi) (MB)

4

5.5

6

FL Communication

(θg=⊕icθi (MB)

2.5

3

3.2

Back to article page