Irum Hirra
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Irum Hirra has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Ahmed Mohammed Alghamdi
Department of Software Engineering, College of Computer Science and Engineering, University of Jeddah, Jeddah, Saudi Arabia
1 shared publication - Ahmed S. Alfakeeh
Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia
1 shared publication - Ayaz Hussain
Advanced Network Architectures Lab (CRAAX), UPC BarcelonaTech, 08800 Vilanova, Spain
1 shared publication - Iftikhar Ahmed Saeed
Department of Computer Science and IT, The University of Lahore, Sargodha Campus, Sargodha, Pakistan
1 shared publication M. Usman Ashraf
Department of Computer Science, University of Management and Technology Sialkot, Sialkot, Pakistan
1 shared publication- Mubashir Ahmad
IRC for Finance and Digital Economy, King Fahd University of Petroleum & Minerals, Dhahran, Saudi Arabia
1 shared publication
Latest Publications
- Top Cited PapersOpen Access
- 2 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 24273-24287
Breast Cancer Classification From Histopathological Images Using Patch-Based Deep Learning Modeling
Accurate detection and classification of breast cancer is a critical task in medical imaging due to the complexity of breast tissues. Due to automatic feature extraction ability, deep learning methods have been successfully applied in different areas, especially in the field of medical imaging. In this study, a novel patch-based deep learning method called Pa-DBN-BC is proposed to detect and classify breast cancer on histopathology images using the Deep Belief Network (DBN). Features are extracted through an unsupervised pre-training and supervised fine-tuning phase. The network automatically extracts features from image patches. Logistic regression is used to classify the patches from histopathology images. The features extracted from the patches are fed to the model as input and the model presents the result as a probability matrix as either a positive sample (cancer) or a negative sample (background). The proposed model is trained and tested on the whole slide histopathology image dataset having images from four different data cohorts and achieved an accuracy of 86%. Consequently, the proposed method is better than the traditional ones, as it automatically learns the best possible features and experimental results show that the model outperformed the previously proposed deep learning methods.
- 2 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 24273-24287
Top Cited PapersOpen AccessBreast Cancer Classification From Histopathological Images Using Patch-Based Deep Learning Modeling
Accurate detection and classification of breast cancer is a critical task in medical imaging due to the complexity of breast tissues. Due to automatic feature extraction ability, deep learning methods have been successfully applied in different areas, especially in the field of medical imaging. In this study, a novel patch-based deep learning method called Pa-DBN-BC is proposed to detect and classify breast cancer on histopathology images using the Deep Belief Network (DBN). Features are extracted through an unsupervised pre-training and supervised fine-tuning phase. The network automatically extracts features from image patches. Logistic regression is used to classify the patches from histopathology images. The features extracted from the patches are fed to the model as input and the model presents the result as a probability matrix as either a positive sample (cancer) or a negative sample (background). The proposed model is trained and tested on the whole slide histopathology image dataset having images from four different data cohorts and achieved an accuracy of 86%. Consequently, the proposed method is better than the traditional ones, as it automatically learns the best possible features and experimental results show that the model outperformed the previously proposed deep learning methods.