Page 93 - sjsi

Research Article: Alsibai & Heydari 93

Figure 8: Samples of infected and healthy ovaries in Dataset B

However, to keep the two experiments identical number of ultrasound images which improves
and due to the problem of dataset B being the training process massively.
relatively small in data number, data Dataset B Results
augmentation techniques such as random After training the same model again for 15
horizontal flipping, random vertical flipping, epochs on this new dataset and keeping all the
random brightness alteration, and random hyperparameters fixed, the following results
rotation are applied using the Python library and confusion matrix are obtained:
`imgaug`. Data augmentation increases the

Figure 9: Results of accuracy and loss during training epochs with the new data for Dataset B

Figure 10: Confusion Matrix for Dataset B

SJSI – 2023: VOLUME 1-1