Identification of Drug-Disease Associations Using Information of Molecular Structures and Clinical Symptoms via Deep Convolutional Neural Network

In 2020, researchers developed a deep convolutional neural network model to predict drug-disease associations using molecular structures and clinical symptoms. The model was trained on 20,000 samples and tested on the remaining samples, achieving an accuracy of 89.90% on the training set and 86.51% on the test set. The model was also tested on an independent dataset, where it achieved a prediction accuracy of 73.54%. The method identified 3,620,516 potential drug-disease associations out of 24,266,646 unknown associations. The study concluded that this method could be a powerful tool for new drug research and development.