A team led by Prof. Rabiee at AI-Med has designed and implemented the AI-based COVID-19 Detection system which recognizes anomalies that are visually hard to detect in chest CT Scan images immediately after a patient is infected with COVID-19.

The model utilizes an innovative preprocessing module to remove possible batch effects and deploys data from multiple sources for end-to-end training of Deep Neural Networks (DNN). In addition, the model calculates the size and volume of the infected areas for a more effective treatment of COVID-19 patients. The accuracy score of the model is above 95%.


The COVID-19 pandemic continues to evolve at a fast pace worldwide. The AI-Med team at Sharif University of Technology has developed the AI-Med COVID-19 Detection System to enable physicians and other medical experts with prompt and accurate diagnosis measures. This software system utilizes an innovative preprocessing module for batch effect removal along with Interpretable Deep Neural Network (DNN) for rapid and accurate diagnosis of COVID-19. AI-Med uses chest CT Scan images which have been saved as Ground Glass Opacity Axial DICOM standard in a single zip file.

The AI-Med COVID-19 Detection System is free for public medical and research centers. To receive a free desktop copy or API for PACS systems please submit an official request via email or regular mail. We also provide limited access to our online services. Please send us an email with an explanation of the intended use of the online system. Upon approval, we send you the login information.

Email: aimed@ictic.sharif.edu
Address: AI-Med Research Group, Artificial Intelligence Innovation Center, Advanced ICT Research Institute Sharif University of Technology SUT Technology Tower 2 Azadi St., Habibollahi St., Habibzadegan Alley, Fatemi Alley No. 1 East, Unit 11, 4th Floor Tehran, Iran