Table 11 Comparison between our best models and existing work using the same BreaKHis database with the same x40 magnitude of histopathological images and the same ratio for model learning and validation.

From: Enhancing histopathological image classification of invasive ductal carcinoma using hybrid harmonization techniques

Existing work

Ratio

Train/Validation

Recognition rate (%)

Spanhol et al. 2015

70/30

83.90

Sanchez-Morillo et al. 2018

70/30

88.32

Boumaraf et al 2021

70/30

87.69

Our work - model 4

70/30

88.00±3.7

Our work - model 1

70/30

90.15±3.0

Our work - model 2

70/30

92.15±2.2

Our work - model 3

70/30

93.40±1.8

  1. Significant values are in bold.
Back to article page