Earlier Detection of Brain Tumor by Pre-Processing Based on Histogram Equalization with Neural Network

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computer-Aided Diagnosis

Background:

• Magnetic Resonance Imaging (MRI) is crucial for early detection of pathological changes in tissues and organs.
• Medical image segmentation is inherently complex, posing challenges for accurate analysis.
• Accurate segmentation is vital for effective computer-assisted diagnosis (CAD).

Purpose of the Study:

• To develop and evaluate a novel deep learning-based method for medical image segmentation.
• To improve the accuracy of early brain tumor detection using MRI.
• To enhance computer-assisted diagnosis (CAD) systems for neurological conditions.

Main Methods:

• The study utilized deep learning techniques for image processing and segmentation.
• A preprocessing step involved adaptive histogram contrast normalization with learning-based neural quantization (AHCN-LNQ).
• The proposed method was applied to MRI scans for brain tumor identification.

Main Results:

• The proposed AHCN-LNQ method achieved a high accuracy of 93% in detecting brain tumors.
• The method demonstrated a precision of 92% and a specificity of 94%.
• Simulation outcomes indicate superior performance compared to existing techniques.

Conclusions:

• The developed deep learning approach offers a significant advancement in medical image segmentation for brain tumor detection.
• The AHCN-LNQ method shows promise for improving the reliability and efficiency of CAD systems.
• This technique has the potential to aid in the early and accurate diagnosis of neurological abnormalities using MRI.