This study proposes an enhanced image analysis system for gastric cancer using convolutional neural networks (CNNs). The system aims to improve the accuracy and efficiency of gastric cancer diagnosis by automatically learning and extracting meaningful features from gastric cancer images, without requiring manual feature engineering. The proposed methodology for feature extraction and classification includes CNNs and transfer learning. The Kaggle dataset is used to train and evaluate the performance of the system. The result shows that the proposed system can achieve high accuracy 70% and efficiency in gastric cancer diagnosis, with the potential to reduce inter-observer variability and enable early detection and treatment. However, it is important to note that the system should always be used in conjunction with clinical expertise and judgment and should not replace the role of trained medical professionals in the diagnosis and treatment of gastric cancer.