COVID-19 Projects from CSA, IISc
Project 1: Lockdown and Other Policies for Containing COVID-19 in Small Worlds
Authors: V. Vinay (Ati Motors) and C. Bhattacharyya (CSA, IISc)
Contact (left to right): vinay@atimotors.com, chiru@iisc.ac.in
Our aim in this project is to understand what policies can be implemented after lockdown.
Small-world models are useful tools in network epidemiology. A city consists of many wards. We model such cities as a Multi-Lattice Small World (MLSW) network, where each ward of a city is modelled as a 2D lattice and nearby wards are connected together. We simulate several interventions on MLSW and study their effectiveness in suppressing COVID-19 on such networks. Our study highlights three findings:
Usual contact tracing involves tracing the immediate contacts. If that can be enhanced to tracing the contacts and their contacts, followed by sealing (TC2S), it would have a huge impact.
A restricted work week, such as a 2-day work week, followed by a lockdown can be as effective as a lockdown.
A policy such as ward-wise sealing and opening, depending on the infection levels in the ward, not only has the lowest attack rate (the percentage of the total population infected) but also requires the shortest time for the epidemic to end.
A preliminary draft is available here.
Press coverage: Sealing areas with higher Covid-19 cases or 2-day work week with lockdown can contain virus, shorten epidemic duration: Analysis (Hindustan Times)
Project 2: CovidWATCH — A Rapid COVID-19 Monitoring Tool for Regions with Low Smartphone Penetration
Lead developers: Niharika Venkatesh (AI Foundry) and Nabanita Paul (IISc)
Contact (left to right): niharika@aifoundry.ai, nabanitapaul@iisc.ac.in
Advised by: Arvind Saraf (AI Foundry) and Chiranjib Bhattacharyya (IISc)
Contact (left to right): arvind@aifoundry.ai, chiru@iisc.ac.in
A collaboration between the Indian Institute of Science (IISc) and AI Foundry, Bengaluru
CovidWATCH is a rapid monitoring tool developed for areas with low smartphone penetration.
It offers a basic screening test based on the ICMR strategy and a symptom tracker to record daily symptoms, via a multi-language WhatsApp chatbot. Built specifically for people with little to no technological acumen, it also allows a single volunteer to take the test on behalf of multiple nearby people for convenience. The data is shared with the authorities in the form of a dashboard, which they can filter by location, symptoms, age, etc. for subsequent follow-ups. This tool has been deployed in a ward under the Pune Municipality and has already helped authorities survey close to 3,000 people in about 2 weeks.
A slide deck detailing the tool is available here.
Project 3: COVID-SWIFT (Now Known as XraySetu)
Lead developers: Sabyasachi Sahoo (IISc), Rachit Shah (IISc), Siva Teja Kakileti (Niramai), and Prateek Katte (Niramai)
Contact (left to right): sabyasachis@iisc.ac.in, rshah240@gmail.com, sivateja@niramai.com, pratik.katte@niramai.com
Advised by: Chiranjib Bhattacharyya (IISc), Geetha Manjunath (Niramai), and Dr. Padmanabha Kamath (KMC)
Contact (left to right): chiru@iisc.ac.in, swift@niramai.com
A collaboration between the Indian Institute of Science (IISc), Niramai, and KMC, Bengaluru
COVID-SWIFT is a rapid AI solution for diagnosing COVID-19 from chest X-rays via WhatsApp.
COVID-SWIFT is a free WhatsApp-based service that provides a swift diagnosis of potential COVID-19 patients by analyzing chest X-ray images. Our state-of-the-art deep learning model generates a report containing predictions for COVID-19 and 14 other lung abnormalities, with interpretable semantic markings on the chest X-ray. This can help doctors understand the severity of their patients' illness. We ran a small-scale pilot for the last 10 months, in which interested doctors could receive a machine-generated X-ray report within minutes of sending us chest X-rays of suspected patients. Our model is trained using multi-task learning on multiple chest X-ray datasets from NIH, RSNA, etc. We will soon be sharing our paper with further details.
COVID-SWIFT has now been launched as XraySetu in collaboration between IISc, Niramai, and ARTPARK. XraySetu is quick and simple for busy doctors to use. It allows doctors in rural areas to plan early intervention for their patients by simply taking a picture of the X-ray and sending it over WhatsApp. We believe this could be the model for the future of Indian healthcare, accessible to everyone wherever they might be.
For more information, please visit XraySetu.