A transfer learning approach to implementation of pretrained CNN models for Breast cancer diagnosis

Main Article Content

Rajan Prasad Tripathi, Sunil Kumar Khatri, Darelle Van Greunen Baxodirovna


In this study, different  pre trained model based on CNN for breast cancer picture segmentation and classification has been analysed . Different models including InceptionV3, DenseNet121, ResNet50, VGG16 and MobileNetV2 models are used to categorize Mammographic Image Analysis Society (MIAS), This approach will help radiologist's aide in early recognition and increment the productivity of our framework. Broad exploratory outcome showed the predominant presentation accomplished on account of calibrating a pretrained network. In our study we found out VGG16 and Resnet101 perform better in predicting the cancer classification, VGG16 performs slightly better than Resnet101 in predicting the malignant Tissue. Furthermore, in the study it was found out that VGG 16 and Resnet delivers the greatest accuracy.

Article Details