In this work, Liver cancer segmentation based on Dual-discriminator conditional generative adversarial network optimized with salp swarm Optimization Algorithm is proposed. Initially, CT scan image dataset is collected from Kaggle repository. Then, the CT scan images are pre-processed using the contrast limited adaptive histogram equalization filtering scheme for removing the noises and to enhance the image quality. Then these pre-processed outputs are given to feature extraction methods. In the feature extraction process, empirical wavelet transform method is used. These extracted features are given into Dual-discriminator conditional generative adversarial network for segmenting liver cancer. Then, the hyper parameters of DCGAN classifier is optimized with the salp swarm Optimization Algorithm. The proposed Dual-discriminator conditional generative adversarial network optimized with salp swarm Optimization Algorithm method attains higher accuracy with low execution time.
Liver cancer Dataset (US simulation & segmentation)
Segmented Image
#Liver, #Cancer, #Segmentation, #Simulation, #Discriminator, #Optimization, #Quality, #Dataset, #Kaggle, #Histogram, #Equalization, #Extraction, #Hyperparameters, #Time, #Histopathology, #Diagnosis, #Prognosis, #Bottleneck, #Decoder, #Segmentation, #Detection #Localization, #Pathology, #Maps, #Clinical, #Diagnostics, #Mechanism, #Architecture, #Blocks, # Pathologists
[1] Lal, S., Das, D., Alabhya, K., Kanfade, A., Kumar, A. and Kini, J., 2021. NucleiSegNet: robust deep learning architecture for the nuclei segmentation of liver cancer histopathology images. Computers in Biology and Medicine, 128, p.104075. [2] Kim, Y.J., Jang, H., Lee, K., Park, S., Min, S.G., Hong, C., Park, J.H., Lee, K., Kim, J., Hong, W. and Jung, H., 2021. PAIP 2019: Liver cancer segmentation challenge. Medical Image Analysis, 67, p.101854.