Brain Tumor Classification Using Fine-Tuned GoogLeNet Features and Machine Learning Algorithms: IoMT Enabled CAD...

Area of Science:

• Medical Imaging
• Artificial Intelligence in Healthcare
• Internet of Medical Things (IoMT)

Background:

• The Internet of Medical Things (IoMT) is revolutionizing healthcare by enabling remote patient monitoring and expert consultation.
• Early detection and classification of brain tumors are critical for improving patient survival rates.
• Computer-aided diagnosis (CAD) systems, enhanced by IoMT, offer significant potential in addressing these challenges.

Purpose of the Study:

• To develop and evaluate an IoMT-enabled CAD system for classifying brain tumors.
• To leverage deep learning, specifically Convolutional Neural Networks (CNNs), for enhanced diagnostic accuracy.
• To classify brain tumors into three types: glioma, meningioma, and pituitary.

Main Methods:

• Utilized a transfer learning approach with a pre-trained CNN (GoogLeNet) for feature extraction from brain MRI images.
• Employed classifiers including softmax, Support Vector Machine (SVM), and K-Nearest Neighbor (K-NN) for tumor classification.
• Trained and validated the model on the CE-MRI Figshare and Harvard medical repository datasets.

Main Results:

• The proposed IoMT-enabled CAD system demonstrated superior performance compared to existing models.
• Achieved high accuracy, specificity, and F1 scores in classifying brain tumors.
• Experimental results indicate the model's effectiveness in distinguishing between glioma, meningioma, and pituitary tumors.

Conclusions:

• The developed IoMT-enabled CAD system shows significant promise for accurate and efficient brain tumor classification.
• Deep learning models, particularly CNNs, are highly effective in analyzing medical imaging for early disease detection.
• This approach can contribute to improved patient outcomes through timely diagnosis and treatment planning.