GET IN TOUCH
- Please wait...
The COVID-19 pandemic has forced us to recognize that most countries of the world are underprepared to tackle any major health crisis. Decision-makers have to take decisions that will determine the outcome of the lives of millions of people. Hence, a data-driven approach in fighting Covid-19 may be the need of the hour.
Data will play a crucial role in this process because only the correct data can generate accurate insights for the decision-making process and alleviate the blow of this pandemic.
Data might be used through multiple avenues and channels to perform a wide array of functions that help understand and fight the pandemic better.
Mobility information obtained using aggregated telecom information was used during the Ebola outbreak in West Africa in 2014-2016 and has been further mined by UNICEF and other parties.
In Belgium, Dalberg Data Insights in collaboration with the government has been working to integrate big data into confinement measures.
Anonymized mobility data was harnessed to identify the areas with more movement and infection rates. These areas were put in lockdown step-by-step by using this data. Mobility was reduced by an average of 54% throughout the country.[1]
Due to these prompt data-driven measures, Belgium has brought its death rate down to half and has seen a recovery rate of 67% as of April 13th.[7]
China installed thermal scanners at train stations to detect people with elevated body temperatures. Once detected, citizens were immediately brought in for COVID-19 testing. If the person tested positive, everyone else traveling on that train was notified individually using their national IDs.
Besides using mobile phone data and close circuit cameras, China also rolled out a “Close Contact Detector” app that immediately notified anyone coming in contact with an identified COVID-19 patient.[2]
The South Korean Ministry of Interior and Safety developed an app to stay in touch with patients. The patients can use the app to directly report their symptoms and problems to the responsible authorities.
If a sick person leaves their quarantine zone, both the case officer and the government officials responsible are immediately notified to prevent any further infection.[1]
In-hospital management is also improved by the use of Artificial Intelligence (AI) and Big Data.
“KARMI-BOT” developed by Asimov Robotics is already being trialed for medicine dispensing duties in isolation wards in India. It can also arrange video conferences with caregivers.[3]
Using common factors found in all COVID-19 patients, risk profiles can be developed to understand the areas to direct medical assistance and resources.
Fraym developed a similar map for battling COVID-19 in Nigeria where it took 3 factors into account. The factors were people over age 60, regular smokers, and those who use dirty cooking fuel in their houses.
Machine Learning was employed to better understand the spatial distribution and generate a map.[4]
Big Data is also being used to identify groups in need of assistance and relief. In India, Aadhaar Cards (Biometrically linked national ID cards) are being used to keep track of the neediest and provide them with financial support and food without leakage through intermediaries. In areas that have more Adhaar cards, this has limited food shortages.
This has kept stockpiling and food shortages in check in areas that have more Adhaar Card holders.[5]
The importance of correct data in research is pivotal in ensuring that scientists make the correct observations. With the use of public data sets, scientists can identify trends in infections and decide what factors are responsible for the rapid spread of the disease.
Although this does not provide a certain solution, with enough data inputted and processed, the models tend to be good enough to mitigate some damage.
If data is wrong or incomplete, most statistical models will fail to deliver accurate results. With very low rates of testing all over the globe, authentic and complete datasets are hard to come by. Especially in the case of asymptotic patients, data is hardly ever collected.
Moreover, because of COVID-19’s 14-day incubation period, people start showing symptoms at random times from when they were infected and so the collection of accurate data is very troublesome. Thus, most data-driven approaches in fighting Covid-19 lack universality at the moment and can only be improved with more authentic data.
Another chief reason for data being proven ineffective is a lack of understanding of how the virus works. Scientists are still discovering new mutations of the virus and this has led to a mass misunderstanding about what entails COVID-19 symptoms and what does not. So for the moment, data scientists are being forced to work with the basic symptoms only for data-driven measures.
With extensive tracking by the government, there are legitimate concerns about the privacy of the people, and these measures have faced some criticism.
At MIT, processes are being developed that ensure the privacy of the people while running contact tracing. The process aims to specify only the individuals that would visit a place within a time frame without specifying their movement or their coming and going.[6]
Data collected by Bangladesh is very low because of the country’s lack of proper infrastructure in all parts of the country. Long queues and insufficient testing kits have disrupted proper data collection at every stage. So most decisions have been taken at the general level without much use of big data.
The hotlines for reporting cases are few in number and often very busy due to numerous calls and this makes data collection more difficult.
Most of the solutions implemented in developed countries hinge on the fact that people are connected via some smart device. Given the limited access to technology and lack of technological understanding in the general populace, these solutions would be very difficult to implement in Bangladesh.
The world is still figuring out new ways to fight the pandemic every day and data-driven decisions have been at the heart of this battle.
With most economies on the verge of disaster, workplaces might have to be reconfigured. In that case, the main focus should be on better integration of technology and data in ensuring as least contact as possible in jobs that require physical presence.
Artificial Intelligence can be used to notify at-risk individuals so that they can self-quarantine.
For Bangladesh, the main focus should be collecting more data so that decision-makers can better understand the entire situation.
Besides this, Bangladesh has to make good use of the data at hand, such as area-based infection rates, to make better decisions about the area-wise allocation of direct assistance and resources. All these measures together can help us take better decisions and save lives in the coming days.
Eqra Mohammad Resalat Ohee, Content Writer at LightCastle Partners, has prepared the write-up. For further clarifications, contact here: [email protected]
The LightCastle team has been analyzing the macro and industry level picture and possible impacts wrought about by the Covid-19 crisis. Over the following days, we’ll be covering the major sectors shedding light on the possible short and long term ramifications of the global pandemic. Read all the articles in the series.
Our experts can help you solve your unique challenges
Stay up-to-date with our Thought Leadership and Insights
