Chest diseases are widespread diseases around the world. Pneumonia chest disease is widely spread, especially between children. Medical issues deep learning-based solving becomes trend and more acceptable among both public and medicine societies. Neural Networks are high-level computational algorithms, which prove a high robust performance in image recognition field. Used dataset in this project consist of around 5800 chest x-ray images taken of human patients, divided into two classes normal and affected. Data is split previously with 0.1 ratio. Convolutional Neural Network with adaptive layers is used for 1-dimentional prediction. Results show high performance of the model. The induced accuracy is 92.31 %.