NORMA eResearch @NCI Library

Liver Disease Detection from CT scan images using Deep Learning and Transfer Learning

Jain, Paras (2020) Liver Disease Detection from CT scan images using Deep Learning and Transfer Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (807kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


The liver is an essential organ in the human body and early diagnosis of various liver diseases can be life saving. Computer-aided diagnostic systems can assist doctors in the precise diagnosis of liver diseases and eliminate the invasive process of biopsy. Due to complexities present in CT scan images, images are processed using various transformations, augmentation and segmentation techniques. This research project implements SVM, CNN, Inception-v4 and DenseNet-169 models based on deep-learning and transfer-learning with the aim to develop a binary classifier that can accurately differentiate between healthy or normal and unhealthy or abnormal liver. The different techniques of parameter tuning are employed to improve model performance. Resultantly, it is found that CNN and DenseNet-169 models achieve 98.8 % accuracy and 99.66% accuracy, respectively. However, the computational time of DenseNet-16 is significantly higher than that of the CNN model. A lot more work and research remain, but the current results are promising.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
R Medicine > R Medicine (General)
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 20 Jan 2021 15:30
Last Modified: 20 Jan 2021 15:30

Actions (login required)

View Item View Item