A Novel Approach to Predict Brain Cancerous Tumor Using Transfer Learning

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Brain tumors are a leading cause of cancer mortality with poor survival rates.
• Manual segmentation and classification of brain tumors are challenging, time-consuming, and prone to inaccuracies.
• Distinguishing malignant tumors from normal brain tissue is difficult due to subtle visual similarities.

Purpose of the Study:

• To investigate the efficacy of convolutional neural networks (CNNs) for image-based diagnosis of brain tumors.
• To enhance the accuracy and reliability of brain cancer diagnosis using deep learning.
• To leverage transfer learning for improved classification performance.

Main Methods:

• Utilized CT and X-ray imaging for brain tumor identification.
• Employed Python and Google Colab for the experimental setup.
• Extracted deep features using pre-trained CNN models: VGG19 and MobileNetV2.
• Applied transfer learning to optimize model accuracy.

Main Results:

• Achieved a classification accuracy of 97% using the MobileNetV2 model.
• Obtained a classification accuracy of 91% using the VGG19 model.
• Demonstrated the effectiveness of deep learning models in brain tumor classification.

Conclusions:

• CNNs, particularly MobileNetV2, offer a highly accurate approach for brain tumor diagnosis.
• Early detection through advanced imaging analysis can prevent severe neurological deficits.
• This research supports the integration of AI in oncology for improved patient outcomes.