Table 25 Computational cost vs. performance trade-off analysis.

From: A federated learning-based privacy-preserving image processing framework for brain tumor detection from CT scans

Optimization

Accuracy (%)

F1-score

Training time (sec)

CPU utilization (%)

Memory usage (MB)

Feature reduction (%)

GTO

94.82

0.942

132

71.2

1125

37.4

HBA

95.05

0.948

147

73.5

1180

33.2

ICOA

94.38

0.933

128

69.9

1103

35.6

HGBO

96.42

0.965

176

78.4

1243

28.6

Back to article page