Novel Automatic Segmentation Approach for Early Brain Tumour Detection: Comparative Evaluation with AI Approaches | Chapter 5 | Medical Science: Updates and Prospects Vol. 1

 

Background: Automatic object detection in medical images is a crucial step in the diagnostic process. The problem of detecting brain tumours at an early stage is well advanced with deep learning algorithms (DLA) such as convolutional neural networks (CNN). The issue lies in the fact that these algorithms necessitate a training phase involving a large database of several hundred images, which can be time-consuming and require complex computational infrastructure.

 

Objective: This study aimed to comprehensively evaluate a proposed method, which relies on an active contour algorithm, for identifying and distinguishing brain tumours in magnetic resonance images.

 

Methods: The proposed algorithm was tested using brain images from the BRATS Challenges 2021, specifically focusing on glioma tumours. The proposed segmentation method is made up of an active contour algorithm, an anisotropic diffusion filter for pre-processing, active contour segmentation (Chan-Vese), and morphological operations for segmentation refinement.

 

Results: Its performance was evaluated using various metrics, such as accuracy, precision, sensitivity, specificity, Jaccard index, Dice index, and Hausdorff distance. The proposed method exhibited higher performance measures than most classical image segmentation methods and was comparable to the deep learning methods. These results indicate its ability to detect brain tumours accurately and rapidly.

 

Conclusion: The results section provided both numerical and visual insights into the similarity between segmented and ground truth tumour areas. The findings of this study highlighted the potential of computer-based methods in improving brain tumour identification using magnetic resonance imaging. Future work must validate the efficacy of these segmentation approaches across different brain tumour categories and improve computing efficiency to integrate the technology into potential clinical processes.

 

 

Author(s) Details

Mohammed Almijalli
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

Faten A. Almusayib
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

Ghala F. Albugami
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

Ziyad Aloqalaa
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

Omar Altwijri
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

Ali S. Saad
Department of Biomedical Technology, College of Applied Medical Sciences, King Saud University, Riyadh 11433, Saudi Arabia.

 

 

Please see the book here :- https://doi.org/10.9734/bpi/msup/v1/6551