Back in 2016, DronaMaps was formed with the aim of utilizing AI to map out neighbourhoods, cities and villages in 3D, and eventually use this data for governance and sustainability such as precision agriculture and pipeline planning. Today, the company is enduring the ultimate test of using technology for improved governance, general public health and civic responsibility. They have devised a live dashboard of active COVID19 cases, currently being used by the Govt. of Punjab. In the past few weeks, this dashboard application has been replicated for the Govt of Haryana and four other states.

As students of Johns Hopkins University in Baltimore, USA, founder CEO Utkarsh Singh and cofounder Ayushi Mishra have had sufficient exposure to epidemiology, public healthcare systems and how technology can aid crisis management more efficiently. While Singh set out to start DronaMaps following Uni, Mishra joined him a couple of years later. Together, this duo are bringing a new dimension to analytics and AI using geospatial data applicable to mining, urban planning and now, handling one of the biggest healthcare crises in Indian history.

Tell us about the genesis of DronaMaps and your journey so far.

I studied at IIM Indore, Johns Hopkins University and the University of Oxford, specializing in computer science, data systems and data analysis. I started DronaMaps in 2016 with the aim of utilizing AI for 3D mapping geographical locations. To date, DronaMaps has mapped nearly 4,000 sq.km across six Indian states, working primarily with various government departments and private entities. There is a tremendous amount of data readily available, and with the application of AI, these datasets become very critical in enhancing our understanding of our surroundings, compute it in real time and draw actionable insights.

Tell us about the solution being developed to assist the fight against COVID19

We developed a dashboard that offers multiple ways to track and maintain master patient data on various parameters such as nearest COVID19 hospitals, test center locations, symptomatic individuals, patients admitted & discharged, total number of deaths, samples collected and their results, and total number of admissions in quarantine. It also provides the latitude and longitude of patients through the mobile app with live GPS tracking. There are multiple modalities of tracking, with all data fused together for higher accuracies. This is done by tracking through the phone app using live GPS, tower CDR data, and cell operator VLR data. This data can be analysed for better insights including the trend line, rate of increase and rate of increment.

Even specific and significant milestones in patient data such as death rates could also be deduced. Additionally, heat maps and red zoning of districts and sub-district neighbourhoods based on real time quarantine data as well as geofencing and alerts for high risk patients can be done. Geofencing is a particularly useful feature for at-risk individuals and those showing symptoms. The tool allows for real time data to be received from cell towers and dynamically set geofencing for suspected cases (although this could require a lot of tower data). In addition, we have a citizen-centric mobile webpage to track a person’s location in red zones, quarantine and curfew areas, and can be done using progressive web & mobile apps with platforms that allow COVID19 solutions, websites and codes.

You recently launched an app – tell us about the app and its features.

Yes, we launched the COVA app which allows an admin portal login and live tracking of patients, provides public portal and mobile dashboards, displays information on the nearest testing centers and COVID19 treatment centres based on current location, provides the latest Govt advisories and displays reports of any mass gathering nearby.

Based on your first project in Punjab, what are your key learnings?

Given that we were among the earliest developers of a live dashboard tracking an outbreak in real time, we discovered many challenges such as the nature of streamlined data that has to flow from healthcare workers on the ground, to various technical departments like radiology for instance, and finally administrative departments. Every government department has their own depth of data, and ways of gathering them. We’re still working on optimizing data collection, not just to track on public and administrative dashboards but also to maintain consistency with a possible increase in data loads. An optimized process of data gathering would also help with faster and accurate predictive analysis as well.

This is an unprecedented situation, so we can anticipate lack of preparedness in geospatial infrastructure in certain states to handle this kind of data collection. The necessary infrastructure is in fact, being created real time while product development takes place simultaneously.

We’re also coming across case studies from other nations that are utilizing drones for major operations like delivery, surveillance and disinfectant spraying. Since we work extensively with drones for our regular operations, we’re working on preparing our network, should the need arise.

How has the deployment been so far?

Currently, we are conducting the spectrum of analysis of the all data, of all the patients in Rajasthan. This requires us to work with the Chief Secretary, PSIT, IT Commissioner and District Magistrates, sending them analysis reports and areas to quarantine every hour. Rajasthan is the most successful in terms of administrative success and analysis. I’d say Punjab would be second. However, we have a huge user base in Punjab of over 28 lakh. Haryana Govt mainly uses our admin portals, but we are not doing patient movement tracking and analysis yet. Chattisgarh too is at the same stage. We are adding states more as we speak. UP is next, followed by Kerala, Maharashtra, and hopefully Gujarat.

Clearly, this is a scalable model given your rapid deployment with other states. Can you verify?

Currently, we are leveraging our collective expertise in geospatial analytics to crunch data and aid the government in their fight against COVID19. When we first began working with the Punjab govt, we gathered that this model needs to be scalable so that other states can implement them without losing valuable time in upgrades and fixes. Most states right now are looking for granular tracking of the cases and support in deploying the technology that enables the same. We replicated this model with Haryana and are currently at varying levels of implementation with four other states.

Image Source: DronaMaps