Gastric cancer (GC) is one of the leading causes of mortality from cancer around the world. It primarily affects older persons. Every year, almost six out of ten persons diagnosed with stomach cancer are above the age of 65. This paper proposed a novel GAstric cancer Detection using DEep LEarning (GADDLE) to identify the gastric cancer in an CT image. Initially the input images are pre-processed using the Gaussian adaptive bilateral filter to enhance the quality of the image. Therefore, the pre-processed images are fed into RegNet feature extraction model to for extracting the features in the image. The best features are selected by using the Dingo Optimization Algorithm. Finally, the normal and the abnormal case of the GC are classified using Link Net model. The GasHisSDB datasets are used to evaluate the performance of the proposed method of specific matrices such as Specificity, Recall, Accuracy, Precision and F1-Score. The suggested Link Net improves accuracy by 2.43%, 4.89%, and 1.98% compared to Alex Net, Google Net, and ResNet, respectively