Dense Tissue Pattern Characterization Using Deep Neural Network Academic Article uri icon

abstract

  • AbstractBreast tumors are from the common infections among women around the world. Classifying the various types of breast tumors contribute to treating breast tumors more efficiently. However, this classification task is often hindered by dense tissue patterns captured in mammograms. The present study has been proposed a dense tissue pattern characterization framework using deep neural network. A total of 322 mammograms belonging to the mini-MIAS dataset and 4880 mammograms from DDSM dataset have been taken, and an ROI of fixed size 224 × 224 pixels from each mammogram has been extracted. In this work, tedious experimentation has been executed using different combinations of training and testing sets using different activation function withAlexNet,ResNet-18model. Data augmentation has been used to create a similar type of virtual image for proper training of the DL model. After that, the testing set is applied on the trained model to validate the proposed model. During experiments, four different activation functions ‘sigmoid’, ‘tanh’, ‘ReLu’, and ‘leakyReLu’ are used, and the outcome for each function has been reported. It has been found that activation function ‘ReLu’ perform always outstanding with respect to others. For each experiment, classification accuracy and kappa coefficient have been computed. The obtained accuracy and kappa value for MIAS dataset usingResNet-18model is 91.3% and 0.803, respectively. For DDSM dataset, the accuracy of 92.3% and kappa coefficient value of 0.846 are achieved. After the combination of both dataset images, the achieved accuracy is 91.9%, and kappa coefficient value is 0.839 usingResNet-18model. Finally, it has been concluded that theResNet-18model andReLuactivation function yield outstanding performance for the task.

authors

  • Kumar, Indrajeet
  • Kumar, Abhishek
  • Kumar, V D Ambeth
  • Kannan, Ramani
  • Vimal, Vrince
  • Singh, Kamred Udham
  • Mahmud, Mufti

publication date

  • 2022

number of pages

  • 23

start page

  • 1728

end page

  • 1751

volume

  • 14

issue

  • 5