NORMA eResearch @NCI Library

Brain Age Classification from Brain MRI using ConvCaps Framework

Kumar, Animesh (2020) Brain Age Classification from Brain MRI using ConvCaps Framework. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (3MB) | Preview


Predicting brain age from brain Magnetic Resonance Imaging (MRI) can play a vital role in identifying various neurological disorders at an early stage. Change in brain contour is a great biomarker for onset of brain related problems. Artificial Intelligence has proven its applicability for image classification. However, higher complexity of the architecture and computational overhead are some of the reasons holding its application in actual medical scenarios. Use of conventional CNN has several pitfalls like positional invariance among the features and over-fitting with deeper network architectures. This study encompasses the application of novel Convolutional Capsule network, to inspect relevancy of spatial features and positional invariance while classifying brain age from brain MRI. Transfer learning based models like InceptionV3 and DenseNet were also considered for comparison an analysis. Models are trained on the OASIS (Open Access Series of Imaging Studies) dataset having 436 different brain MRI images and evaluated using model accuracy. In addition, benchmarks like Precision, Recall and F1-Scores were also applied. ConvCaps architecture reached an accuracy of 81% whereas InceptionV3 was slightly better with 85% accuracy. Both models have shown promising results for brain age classification and can be tuned for wider application.

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
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 18 Jan 2021 15:23
Last Modified: 18 Jan 2021 15:23

Actions (login required)

View Item View Item