Probable Region Identification and segmentation in Breast Cancer using the DL-CNN
| Autor: | Anand Jatti, M Nagendra Kumar, C K Narayanappa |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
medicine.diagnostic_test
Computer science business.industry Pattern recognition medicine.disease Convolutional neural network Image (mathematics) Reduction (complexity) Identification (information) Breast cancer medicine Mammography Segmentation Artificial intelligence Layer (object-oriented design) business |
| Zdroj: | 2019 International Conference on Smart Systems and Inventive Technology (ICSSIT). |
| DOI: | 10.1109/icssit46314.2019.8987818 |
| Popis: | Breast Cancer in women is one of the most diagnosed diseases and it is one of the leading disease, which cause death. In past several research works have proposed various methodology to detect the cancer, however due to the Complex nature of micro calcification as well as masses it has the complex nature. Hence in this paper we have proposed a CNN based methodology named Dual layer CNN(DL-CNN), where we have used two layer Convolution Neural Network, first layer Is used for the Probable Region Identification and second layer is used for the Segmentation and false positive reduction. DL-CNN technique is robust in nature and identify the region in efficient manner. Moreover, for the evaluation we have used In Breast image dataset, other parameter considered are True Positive Rate at False positive per image. DL-CNN scores 0.9726 at 0.39706 respectively, it outperforms when compared to the other existing technique. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|