This document presents a convolutional neural network model for automatic dermatology disease diagnosis. It uses the HAM10000 public dataset of dermatoscopic images which contains over 10,000 images across 7 common dermatological diseases. The proposed CNN model contains convolutional and max pooling layers to extract features from the images, followed by fully connected layers for classification. Data augmentation is applied to increase training samples and reduce overfitting. The model achieves an accuracy of 82% on the HAM10000 test dataset after 20 epochs of training, demonstrating the ability of CNNs to classify skin diseases from images.