الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 166 |  |  |
| | | from | 166 | | |
Abstract
Cancer is a serious public health problem worldwide, presenting high mortality rates when not properly diagnosed and treated and overloading both public and private health systems. Despite the significant progress reached through diagnostic imaging technologies, the death rate is still high. integrating ocular examination to microscope images of histology. Histopathological analysis is a highly skilled and time-consuming process, heavily reliant on the pathologists’ experience, and significantly affected by elements like exhaustion and drowsiness.
It begins with an introduction to deep learning as an efficient technology and why we need it in our lives, then introduces the outline and objectives of the research.
Next, Feature engineering, a challenging and time-consuming procedure that employs prior expert domain knowledge of the data to build effective features, is discussed in the thesis as a key factor in the success of most conventional classification algorithms. By contributing, this gap can be eliminated.
The current study includes ISIC 2019, a publicly accessible image collection that includes histological pictures of skin cancer. This work also offers a different method for categorizing these challenging photos without using any explicit segmentation. Such a method tackles the various case learning approach in addition to automatic modeling, particularly employing convolutional neural networks.
Using intelligent techniques such as neural networks, fusion deep features, and interval neutrosophic sets, these policies achieve efficient resource usage and improve the overall performance.
The collected experimental findings have shown that this concept is feasible and have provided guidelines for improving such a model.
The research also presents a comparison between the latest clustering systems in the field of breast cancer diagnosis using available WDBC data.
The results showed the feasibility and preference of the selected research method as an effective technique used to improve the rate of cancer classification and how it can contribute to saving lives.
Finally, the study presents a comparison of results between the introduced policies and the existing policies to assess the proposed model enhancement.
After applying the suggested framework, the accuracy of Fused GoogleNet and Fused DarkNet was enhanced to 82 and 85%, respectively.
Applying SVNSs changed accuracy from 84 to 86%. This demonstrates that it might be obvious that using NS approaches boosts accuracy.
Cancer is a serious public health problem worldwide, presenting high mortality rates when not properly diagnosed and treated and overloading both public and private health systems. Despite the significant progress reached through diagnostic imaging technologies, the death rate is still high. integrating ocular examination to microscope images of histology. Histopathological analysis is a highly skilled and time-consuming process, heavily reliant on the pathologists’ experience, and significantly affected by elements like exhaustion and drowsiness.
It begins with an introduction to deep learning as an efficient technology and why we need it in our lives, then introduces the outline and objectives of the research.
Next, Feature engineering, a challenging and time-consuming procedure that employs prior expert domain knowledge of the data to build effective features, is discussed in the thesis as a key factor in the success of most conventional classification algorithms. By contributing, this gap can be eliminated.
The current study includes ISIC 2019, a publicly accessible image collection that includes histological pictures of skin cancer. This work also offers a different method for categorizing these challenging photos without using any explicit segmentation. Such a method tackles the various case learning approach in addition to automatic modeling, particularly employing convolutional neural networks.
Using intelligent techniques such as neural networks, fusion deep features, and interval neutrosophic sets, these policies achieve efficient resource usage and improve the overall performance.
The collected experimental findings have shown that this concept is feasible and have provided guidelines for improving such a model.
The research also presents a comparison between the latest clustering systems in the field of breast cancer diagnosis using available WDBC data.
The results showed the feasibility and preference of the selected research method as an effective technique used to improve the rate of cancer classification and how it can contribute to saving lives.
Finally, the study presents a comparison of results between the introduced policies and the existing policies to assess the proposed model enhancement.
After applying the suggested framework, the accuracy of Fused GoogleNet and Fused DarkNet was enhanced to 82 and 85%, respectively.
Applying SVNSs changed accuracy from 84 to 86%. This demonstrates that it might be obvious that using NS approaches boosts accuracy.