Brain tumor segmentation and classification using MRI: Modified segnet model and hybrid deep learning architecture...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Brain tumors are a leading cause of death globally, with diagnosis delays increasing mortality.
• Current manual interpretation of Magnetic Resonance Imaging (MRI) for brain tumor detection is time-consuming and prone to errors.
• Radiologist expertise is crucial but limited in traditional diagnostic methods.

Purpose of the Study:

• To develop an automated approach for brain tumor segmentation and classification (BTS&C) using MRI.
• To enhance the accuracy and efficiency of brain tumor diagnosis.
• To overcome the limitations of manual MRI interpretation.

Main Methods:

• Image fusion of T1, TIC, T2, and T2 FLAIR MRI sequences using an improved fusion technique.
• Preprocessing of fused images with Median Filtering (MF).
• Segmentation using a Modified Segnet model with a novel pooling operation.
• Feature extraction including Improved local Gabor Binay pattern Histogram Sequence (ILGBPHS), Weber Local descriptor (WLD), and Tetrolet waveform.
• Classification employing HDLA, a hybrid model combining Bi-LSTM and Modified Linknet.

Main Results:

• The proposed BTS&C method achieved a high accuracy of 98% at 90% detection threshold (TD).
• Demonstrated superior performance compared to established methods like Bi-LSTM, Link Net, LeNet, Squeeze Net, Efficient Net, HHOCNN, and CNN-SVM.
• Indicated significant improvements in diagnostic accuracy and efficiency.

Conclusions:

• The developed automated BTS&C system offers a promising solution for accurate and timely brain tumor diagnosis.
• The proposed method effectively addresses the challenges associated with manual MRI interpretation.
• This AI-driven approach has the potential to improve patient outcomes by reducing diagnostic delays.