Reinventing disease impact simulation models with BharatSim

Chhaya Yadav, Portfolio Head, Engineering for Research, Thoughtworks and Jayanta Kshirsagar, Lead Developer, Engineering for Research, Thoughtworks highlights that for a country as populous and diverse as India, it is imperative to have a simulation framework that can track disease spread, test outcomes in simulations and aid in devising policy interventions

The COVID-19 pandemic brought to light the critical need for advanced technological implementation that could work as a predictive asset when analysing a disease or pandemic’s impact on the general populace. While there are widely applied research methodologies that can gauge the impact of a pandemic and other diseases, many such methodologies have not been able to offer acutely accurate predictions. The lack of agent based simulations has often been a factor that prevented researchers from predicting outcomes and this often hindered governments from formulating efficient policies and regulations to prevent and manage the spread of infectious diseases.

Imagine a scenario where a medical research organisation is studying the effects of a particular disease. They will want to understand the disease’s scope, how it impacts the whole population across different communities with near-accurate data on outcomes and the policy interventions that the government can roll out. Understanding a pandemic or disease’s impact at such a wide scale is a mammoth task – read, cost and time it takes to collect data and accurately replicate an entire population.

While accurate disease prediction seemed like a far-fetched dream a while ago, today, it is possible with models such as BharatSim, which can assess the impact of a pandemic or infectious disease at the most granular level, providing unprecedented levels of detail.

For a country as populous and diverse as India, it is imperative to have a simulation framework that can track disease spread, test outcomes in simulations and aid in devising policy interventions. Such an advanced simulation mechanism would be able to properly predict outcomes while including the country’s extreme levels of social disparity, economic disparity. This need of the hour that could take into account the complexity of Indian cities and the complexities of social interactions resulted in the formulation of BharatSim.

This collaborative project formulated under the guidance of Dr Gautam Menon, professor at Ashoka University and funded by the Bill and Melinda Gates Foundation, is India’s first ultra-large-scale agent-based simulation framework that acts as a disease-modeling system by predicting disease progression, thereby enabling targeted government healthcare interventions.

Initially launched to chart the spread of the COVID-19 pandemic, BharatSim today can operate as an open-source platform that helps Indian researchers formulate strategies to manage transmissible and non-transmissible diseases. BharatSim follows an agent-based modeling system that represents how individuals interact and transmit diseases amongst themselves. The framework alters collective data in real-time and provides ways of understanding and analyzing the epidemic spread in a way that other existing models are not capable of.

The model explores potential impact of a lethal variant, incorporates a high level of detail and flexibility, compares different interventions and provides a powerful tool for decision making. Comparing predictions with data in real-time allows health experts to stay ahead of the disease’s spread, even at early stages where only little about a new variant is known.

The framework also takes into account attributes such as age, weight and geographical region when creating a simulation, which is often not feasible considering the mass scale of the population and their individual characteristics. This is especially crucial to researchers considering pandemics like the COVID-19 affects people on an individualistic level, depending on their personal characteristics and prevalent health conditions. With BharatSim, researchers are now able to predict outcomes with better accuracy.

The simulation framework comprises two components – the simulation engine and the visualisation engine. The simulation engine has been designed to create a synthetic population that is reflective of India’s population. It incorporates daily behaviors and governmental interventions that may affect the management of diseases. Through the visualisation engine’s visual cues, researchers can simultaneously view multiple results and outcomes. This reduces the time taken for a detailed analysis of the simulation and the time taken to present the collected data in an organised and systematic manner.

The analytical data that is a result of leveraging such a simulation model can be a substantial support for governments’ health departments. During the COVID-19 pandemic, health departments across the world were constantly impeded by uncertainty and lack of information regarding the spread, progression and evolution of the virus. This slowed the formulation of effective policies that could tackle the pandemic and safeguard citizens. However, with the establishment of BharatSim, governments will be able to access and utilise near-accurate real time data to help predict the social and economic outcomes of a disease, enabling them to formulate better interventions.

Advancement of research technology goes a long way towards establishing a structured plan of action, and can be incredibly beneficial, not just for those within the scientific community, but for governments and in turn, the citizens of a country as well.

Data analysisdigital healthdisease management
Comments (0)
Add Comment