Enhanced Liver Tumor Detection Using Deep Learning Techniques for Biomedical Image Segmentation

Gurazada, Vigneswara Venkata Sai Nilesh (2024) Enhanced Liver Tumor Detection Using Deep Learning Techniques for Biomedical Image Segmentation. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (887kB) | Preview

Abstract

Liver tumor segmentation is a fundamental step in medical imaging, and is of paramount importance to liver cancer diagnosis and treatment planning. In this study, we introduce LT-Net (Liver Tumor Network), a deep learning model that is able to segment liver tumors from CT and MRI scans, entirely automatically. In particular, LT-Net uses a novel architecture consisting of parallel convolutional layers in the encoder, upsampling in the decoder, and ResNet50 as a backbone, to improve tumor detection accuracy. A diverse dataset is used to train and evaluate the model, achieving a Dice Coefficient of 0.9733, IoU Score of 0.9705 and Accuracy of 0.9986. The model also shows a PSNR of 25.44, which demonstrates that it is capable of preserving fine details while properly segmenting tumors. Evaluation studies indicate that LT-Net is highly effective in identifying liver tumors and the performance is superior to conventional approaches and has promising adaptations for real-time use in clinic. To rectify these problems, future work will be to increase the segmentation accuracy, employ multi-modal imaging, and optimize for real-time application to assess generalizability across different patient cohorts. According to the results of the LT-Net model, it can be inferred that it is highly effective for improving the diagnostic accuracy and treatment speed of liver cancer to provide timely decision-making references for clinicians.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Tomer, Vikas UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > Biomedical engineering R Medicine > Healthcare Industry Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning Q Science > Life sciences > Medical sciences > Pathology > Tumors
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	02 Sep 2025 12:16
Last Modified:	02 Sep 2025 14:42
URI:	https://norma.ncirl.ie/id/eprint/8706

Actions (login required)

View Item