Tools
Tools
Pereppadan Ignatious, Sandra (2023) Liver Segmentation on CT Images for Tumor Detection Using Hybrid Modelling of U-Net, Inceptionv3 and ResNet18. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (885kB) | Preview |
Preview |
PDF (Configuration manual)
Download (773kB) | Preview |
Abstract
The accurate Diagnosis of disease is highly important in terms of patient treatment. Now a days the tumors like brain tumors, liver tumors etc. are getting higher, where an early and accurate detection will help the patients in treatment. Hence the Deep learning methods can be applied to this scenario where an accurate liver segmentation and tumor detection can be done. My project proposes the solution where a hybrid modelling is being done for the liver tumor detection with U-net, Inceptionv3 and Residual Neural Network (ResNet18). The Research Institute Against Digestive Cancer (IRCAD) dataset contains images of 20 different patients including both genders with 75%of them having tumor. The data is hence used to apply the deep learning methods to apply the segmentation. The hybrid modelling has been applied to the data, where U-Net is the backbone model. While application of U-Net + Inceptionv3 and U-Net + ResNet18, UNet+ResNet18 performs the best. The project is applied the grey wolf optimiser (GSO) and particle swarm optimiser (PSO) in the preprocessing for the selected model of ResNet18 where the grey wolf optimiser performed the best with average accuracy of 99% and it also have F1 score of 0.98 which also shows that the model has good performance. Furthermore, the model can be useful for the radiotherapy process for cancer treatment and can be integrated with other medical systems for an automatic treatment procedure.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Makki, Ahmed UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > RB Pathology R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer) Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 28 Dec 2024 14:26 |
| Last Modified: | 28 Dec 2024 14:26 |
| URI: | https://norma.ncirl.ie/id/eprint/7248 |
Actions (login required)
 |
View Item |