A robust grey wolf-based deep learning for brain tumour detection in MR images

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• Brain tumor detection is crucial, but current image processing methods face time challenges.
• Accurate and efficient detection models are needed to overcome these limitations.

Purpose of the Study:

• To propose a novel and accurate brain tumor detection model.
• To improve the efficiency and performance of existing detection techniques.

Main Methods:

• The proposed GW-DBN model integrates preprocessing (contrast enhancement, skull stripping), fuzzy c-means clustering (FCM) segmentation, and feature extraction using GLCM and GRLM.
• Classification is performed using an optimized deep belief network (DBN) enhanced with grey wolf optimization (GWO).

Main Results:

• The GW-DBN model demonstrated superior performance compared to conventional methods.
• Evaluated metrics include accuracy, specificity, sensitivity, precision, NPV, F1-Score, MCC, FNR, FPR, and FDR, highlighting the model's effectiveness.

Conclusions:

• The developed GW-DBN model offers a significant advancement in brain tumor detection accuracy and efficiency.
• This optimized approach holds promise for improving diagnostic outcomes in neuro-oncology.