Artificial Neural Network for Classifying Injected Materials under Ultrasonography

Galuh Retno Utari (Universitas Kristen Satya Wacana)
Giner Maslebu (Universitas Kristen Satya Wacana)
Suryasatriya Trihandaru (Universitas Kristen Satya Wacana)

Advance Sustainable Science, Engineering and Technology (ASSET)
Vol 3, No 1 (2021)

We have constructed an artificial neural network (ANN) architecture to classify four different classes of ultrasonography recorded from a jelly box phantom that was injected by iron, glass, or plastic marble, or without any injection. This jelly box was made as a phantom of a human body, and the injected materials were the cancers. The small size of the injected materials caused only little disturbances those could not easily distinguished by human eyes. Therefore, ANN was used for classifying the different kind of the injected materials. The number of original imagestaken from ultrasonographs were not so many, therefore we did data augmentation for providing large enough dataset that fed into ANN. The data augmentation was constructed by pixel shifting in horizontal and vertical directions. The procedure proposed here produced 98.2% accuracy for predicting test dataset, though the result was sensitive to the choice of augmentation area.