Compressed Deep Learning and Transfer Learning Model for Detecting Brain Tumour

The accurate detection of brain tumors is of paramount importance in medical imaging for early diagnosis and effective treatment planning. In this study, we explore the application of deep learning and transfer learning techniques to detect brain tumors using medical images. Our proposed model incorporates VGG16, VGG19, and EfficientNetB3, with EfficientNetB3 utilized as the transfer learning base. Remarkably, our model achieves an impressive accuracy of 92% for brain tumor detection. In this article a genetic algorithm-based compression approach is used to reduce model size and computational demands, while ensuring accuracy is preserved. The outcomes demonstrate substantial reductions in storage space and improved inference time. Our literature review underscores the efficacy of deep learning and transfer learning in brain tumor detection, indicating promising applications in healthcare. Overall, this research contributes to the advancement of medical imaging and brain tumor diagnosis, presenting a valuable tool for medical professionals in their endeavors to combat this life-threatening disease.