Author:
Haq Amin ul,Li Jian Ping,Khan Shakir,Alshara Mohammed Ali,Alotaibi Reemiah Muneer,Mawuli CobbinahBernard
Abstract
AbstractThe classification of brain tumors (BT) is significantly essential for the diagnosis of Brian cancer (BC) in IoT-healthcare systems. Artificial intelligence (AI) techniques based on Computer aided diagnostic systems (CADS) are mostly used for the accurate detection of brain cancer. However, due to the inaccuracy of artificial diagnostic systems, medical professionals are not effectively incorporating them into the diagnosis process of Brain Cancer. In this research study, we proposed a robust brain tumor classification method using Deep Learning (DL) techniques to address the lack of accuracy issue in existing artificial diagnosis systems. In the design of the proposed approach, an improved convolution neural network (CNN) is used to classify brain tumors employing brain magnetic resonance (MR) image data. The model classification performance has improved by incorporating data augmentation and transfer learning methods. The results confirmed that the model obtained high accuracy compared to the baseline models. Based on high predictive results we suggest the proposed model for brain cancer diagnosis in IoT-healthcare systems.
Funder
National Natural Science Foundation of China
Publisher
Springer Science and Business Media LLC
Reference32 articles.
1. Roser, M. & Ritchie, H. Cancer. Our World in Data (2015). https://ourworldindata.org/cancer.
2. Khan, H. A., Jue, W., Mushtaq, M. & Mushtaq, M. U. Brain tumor classification in MRI image using convolutional neural network. Math. Biosci. Eng 17, 6203–6216 (2020).
3. Swati, Z. N. K. et al. Content-based brain tumor retrieval for MR images using transfer learning. IEEE Access 7, 17809–17822 (2019).
4. Sultan, H. H., Salem, N. M. & Al-Atabany, W. Multi-classification of brain tumor images using deep neural network. IEEE Access 7, 69215–69225 (2019).
5. Bishop, C. M. Pattern Recognition and Machine Learning (Springer, 2006).
Cited by
31 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献