Brain tumour detection systems based on histopathological image analysis using segmentation and classification by deep learning architectures
by Abdullah O. Alamoudi
International Journal of System of Systems Engineering (IJSSE), Vol. 14, No. 5, 2024

Abstract: The human brain has billions of cells and is one of the body's most complicated organs. This research proposes a novel technique in histopathological image analysis for detecting brain tumour by segmentation with classification utilising deep learning (DL) methods. Input images have been taken as histopathological images and processed for noise removal, smoothening, and normalisation using adaptive median filtering and MachenoStain normalisation. Then the processed image was segmented utilising an active contour-based Kernel k-means clustering operation where the tumour region has been segmented, and this segmented part has been classified. Boltzmann Q-learning with a convolutional network classified the tumour region and analysed its volume. Experimental analysis is carried out for various histopathological brain images for the proposed technique compared to the existing technique. The parameters compared are accuracy of 96%, sensitivity of 91%, specificity of 86%, coefficient of dice of 85%, Jaccard's coefficient of 96%, spatial overlap of 68%, AVME of 53%, and FoM of 63%.

Online publication date: Mon, 02-Sep-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.

Pay per view:
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 System of Systems Engineering (IJSSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your 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