PBTNet: A New Computer-Aided Diagnosis System for Detecting Primary Brain Tumors

Area of Science:

• Neurology
• Computer Science
• Medical Imaging

Background:

• Brain tumors are a significant cause of mortality, with over 120 types categorized as primary or metastatic.
• Early and accurate diagnosis of primary brain tumors is crucial for effective treatment.
• Manual interpretation of magnetic resonance imaging (MRI) for brain tumor detection has high inter-observer variability.

Purpose of the Study:

• To introduce PBTNet, a novel computer-aided diagnosis system for detecting primary brain tumors in MRI scans.
• To address the limitations of subjective manual interpretation in brain tumor diagnosis.

Main Methods:

• PBTNet utilizes a pre-trained ResNet-18 model, fine-tuned for feature extraction.
• Three randomized neural networks (Schmidt neural network, random vector functional-link, extreme learning machine) act as classifiers.
• An ensemble approach combines the outputs of the three classifiers for final predictions.

Main Results:

• The PBTNet system demonstrated effectiveness in diagnosing primary brain tumors.
• 5-fold cross-validation confirmed the classification performance of the proposed system.

Conclusions:

• The developed PBTNet system is a promising tool for the accurate and efficient diagnosis of primary brain tumors.
• The computer-aided approach can potentially reduce diagnostic variability and improve patient management.