A new machine learning approach to classify MRI of brain tumour using SAE + LSTM
by Biswaranjan Mishra; Kakita Murali Gopal; Bijay Kumar Paikaray; Srikant Patnaik
International Journal of Bioinformatics Research and Applications (IJBRA), Vol. 20, No. 3, 2024
Abstract: A brain tumour is a serious condition that can seriously harm brain cells and eventually progress to cancer, which is life-threatening. The patient's chances of survival can be improved when the tumour stages are detected early. The proposed tumour diagnosis uses a fused feature set to increase the classifier's accuracy. To begin with, the features from the MRI images are extracted using the grey level co-occurrence matrix (GLCM) and histogram of oriented gradients (HOG). After dimensionality reduction, features are chosen with stacked autoencoder (SAE). Second, the high-level features from the MRI images are extracted using the channel-wise attention block. The long short-term memory (LSTM) is trained to produce the results of the classification using the fused features from SAE and the attention block. The proposed approach is evaluated with the BRATS dataset for the years 2018-1020. The accuracy attained over various datasets is 97%, 95.56% and 95.23%.
Online publication date: Mon, 15-Jul-2024
The full text of this article is only available to individual subscribers or to users at subscribing institutions.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Bioinformatics Research and Applications (IJBRA):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com