In India, chest X-rays are given as a standard diagnostic imaging procedure (X-ray CT included) for patients with covid-19 symptoms. The impact of covid-19 on an infected person’s lung is quite severe as reported anecdotally and also in limited studies. Chest X-rays of patients with covid-19 infection show regions of anomalous opacity and can be used as a good indicator of the presence of infection. In addition, quantification of the infection based on infection burden can be used to decide on hospitalization on a patient who tests Covid positive. We propose to develop a deep learning system to detect regions of anomalous opacity in chest X-rays. We will train a deep model using publicly available databases of various lung diseases and employ a few shot learning techniques and other techniques to learn from limited data, using a limited covid-19 chest X-ray images. The system will be trained to classify between anomalous and normal chest X-ray images. It would also be trained to rank severity of the disease based on infection area. The approach to train the network using a few data points will also help to solve the problem of generalization of neural networks to medical imaging datasets, especially across different centers. Any model deployed in small centers can then be improved upon to work well with the image data acquired at that center. The system can be used as a second opinion for diagnosis as well as for hospitalization decisions. In order to ensure that the model does not become defunct once a vaccine becomes available, it will be fine-tuned in accordance with availability of data, to classify some of the common afflictions in the indian population which includes pneumonia and Tuberculosis. A second aim of this work would be to go a step further and automatically delineate infected regions in Chest X-ray CT images, indicative of Covid. This is important as the severity of the disease during treatment is often gauged from X-ray CT images and in many cases CT images are being used for judging if a patient exhibiting symptoms needs hospitalization if he or she tests negative for covid. For X-ray CT analysis we will develop 2D and 3D CNN models to determine the infected regions of the lung. A clinical partner would help provide data for proof of concept