NORMA eResearch @NCI Library

Brain Tumor Detection using Multiple Instance Learning Technique

Chaudhary, Diksha Arvind (2020) Brain Tumor Detection using Multiple Instance Learning Technique. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of dikshaarvindchaudharyconfigurationmanual.pdf]
Download (2MB) | Preview


Brain tumors are the growth of abnormal cells in the brain, which affects millions of people and causes death. Diagnosis of brain tumor along with patient care is crucial. The proposed research aims to identify tumor in brain MRI scans by an automated model. We use a novel method for the detection of brain tumor, which involves multiple instance learning (MIL) based on attention mechanism. Transfer learning based on pre-trained models such as DenseNet121 and InceptionV3 are applied by utilizing BRATS dataset to compare the results. We test the model using evaluation metrics like accuracy, sensitivity, specificity, precision, recall, and F1-score for improving the results. Impressive results are obtained by the proposed system and the results show that MIL is effective in comparison with other models. The results show that the proposed methods study the region of interest by itself and can detect tumors in brain MRI. The result shows that the proposed model is reliable and can be used in the diagnosis of tumor by neuroradiologists for further patient treatment.

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: 22 Jan 2021 11:13
Last Modified: 22 Jan 2021 11:13

Actions (login required)

View Item View Item