Hybrid Deep Learning Model for Brain Tumor Classification Using MRI and CT Images

Brain tumors are a significant threat to human health, making early detection crucial for improving patient survival. Traditional diagnostic methods rely heavily on the expertise of radiologists. Inaccurate identification of tumor types can lead to improper treatment decisions, further complicating patient care. Manual methods, relying on visual inspection of MRI images, are time-consuming and prone to human error, particularly when handling large volumes of complex data. To address these challenges, we propose a hybrid deep learning approach that combines the strengths of Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. This model utilizes both MRI and CT images for tumor classification. CNNs are used to extract deep features from the images, which are subsequently processed by LSTM to capture both spatial patterns and temporal correlations within the data. Our approach works with sequential data, where we process time-series MRI and CT images taken at different time points, as well as image sequences from various angles of the patient. Without using region-based segmentation, this hybrid architecture achieved a training accuracy of ۹۸.۶۰% and a validation accuracy of ۸۲.۱۸%. These results demonstrate the potential of our method in providing accurate and efficient brain tumor classification compared to traditional approaches.

Keywords:

Brain Tumor Classification , Hybrid Deep Learning , CNN , LSTM , MRI/CT Imaging