Early Detection of Brain Tumor using ResNet50 and Transfer Learning

Brain tumor is a disease which can be cause of death of a patient. In India, every year 40000 new brain tumor cases are diagnosed. Identification and early detection of brain tumor will be helpful in timely treating the patient. Manual process of classification of brain tumor suffers from human error. The computer aided detection (CAD) methods are very helpful to reduce this human intervention in brain tumor classification. This research work presents approach of using concept of transfer learning for brain tumor detection. Here we have compared proposed method with various pre trained deep learning models for the classification of brain tumor. We have observed that ResNet50 with modified top layer outperforms over all the available pre trained CNN models.