3DCNN: Three-Layers Deep Convolutional Neural Network Architecture for Breast Cancer Detection using Clinical Image Data
| Autor: | Samad Wali, Amin Ul Haq, Abdus Saboor, Mordecai F. Raji, Wang Zhou, Tao Jiang, Jianping Li, Jalaluddin Khan |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
050101 languages & linguistics
Computer science business.industry Deep learning Model selection 05 social sciences Cancer 02 engineering and technology medicine.disease Machine learning computer.software_genre Convolutional neural network Cross-validation Data set Identification (information) Breast cancer 0202 electrical engineering electronic engineering information engineering medicine 020201 artificial intelligence & image processing 0501 psychology and cognitive sciences Artificial intelligence skin and connective tissue diseases business computer |
| Zdroj: | 2020 17th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP). |
| DOI: | 10.1109/iccwamtip51612.2020.9317312 |
| Popis: | The breast cancer is a critical female disease and its proper identification is very essential for better cure and recovery. The diagnosis of BC is a critical issue for clinical specialists and scholars. Different researchers proposed breast cancer diagnosis methods using deep learning techniques. However, these proposed methods not diagnosis breast cancer accurately. In order to tackle the issue of accurate detection of breast cancer we proposed a 3-layers CNN architecture for accurate detection of breast cancer. The proposed model has been trained and tested on Breast histology images data set. The cross validation method Hold out has been applied for best model selection and hyper parameters tuning. Furthermore, different model evaluation metrics have been used for model performance evaluation. The experimental results demonstrated that propped method is more suitable for breast cancer and it would be incorporated in health care successfully. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|