Arecanut is a tropical fruit also known as betel nut. India is the world's second largest producer and consumer of arecanut. In their life, from the roots to the fruits, they suffer from various diseases. The current method of disease detection is visual inspection, which requires farmers to regularly inspect all crops for disease. In this work, we proposed a system to detect arecanut, leaf and stem diseases using convolutional neural networks and prescribe treatment for them. A convolutional neural network (CNN) is a deep learning algorithm that takes an input as an image, assigns weights and learning scores to various objects in the image, and then learns the results to separate them. To train and test the CNN model, we created our own dataset of 620 healthy and diseased arecanut images. Train data and test data are split in a ratio of 80:20. For model clustering, cross-entropy is used as the loss function, Adam as the optimization function, and accuracy as the measure. They are used 50 times to train the model to achieve high validation and test accuracy with low error. The proposed method was effective in identifying arecanut disease with an accuracy of 88.46%.