Table 2.
Selected articles of brain MRI using deep learning
| Author | Details |
|---|---|
| Ragab et al. [198] | Two-phased segmentation approaches are used for breast tumor segmentation |
| Mambou et al. [199] | Used Independent Component Analysis (ICA) with convolutional neural network to classify breast cancer |
| Selvathi and Poornila [200] | Used Sparse-autoencoder, Stacked Sparse-autoencoder with Convolutional Neural Network for mammogram classification of breast cancer |
| Mohamed et al. [201] | Used multi-fold based technique for breast cancer on biopsy dataset |
| Kavitha et al. [202] | For diagnosing digital mammogram of breast cancer, they used Optimal multi-level Thresholding based segmentation with capsule network |
| Chowdhury et al. [203] | Used transfer learning approach with customized CNN and ResNet101 for classification of breast cancer |
| Escorcia-Gutierrez e al. [204] | Used CNN model with ResNet34 with distinct preprocessing steps for classification of breast cancer |
| Jasti et al. [205] | Used various distinct approaches for feature extraction, selection, image processing and classification of mammograms |
| Jabeen et al. [206] | Used five-fold based deep learning approaches for classification of breast tumor classification from ultrasound |
| Naseem et al. [207] | Used various machine learning based ensemble algorithms to classify breast tumor |
| Singh et al. [208] | Proposed a hybrid approach comprises residual and inception block of CNN for breast cancer classification |
| Liu et al. [209] | Used a pre-trained CNN model AlexNet and fine-tuned on BreakHis, IDC and UCSB datasets |
| Wang et al. [210] | Developed a novel deep learning approach DeepGrad model comprises InceptionV3 blocks for Histopathological image classification |
| Reshma et al. [211] | Used Fourier Transform based Segmentation for Histopathological image (Biopsy) classification of breast cancer |
| Ragab et al. [212] | Proposed Ensembled based deep learning approach containing multi-level thresholding based segmentation for breast tumor |
| Ahmad et al. [213] | Used Gated Recurrent Unit with pretrained CNN model (AlexNet) for classification of Lymph Node of breast tumor |
| Maqsood et al. [214] | Used multi-phase approach comprises contrast enhancement, Transferable texture using pretrained CNN models |
| Ibrokhimov and Kang [215] | Proposed two-stage CNN network to extract local patches from breast cancer and locate Region of Interest (ROI) |
| Mohamed et al. [216] | Used two-step approach, comprises of U-Net (CNN) for extraction fo breast from the whole body and second step is to classify into binary classes |