Science, Art, Litt, Science based Art & Science Communication
Krishna: Scientists create models using various parameters to predict a pandemics’ behaviour and its waves.
The use of machine learning to forecast epidemic spread is a relatively recent advance. Some of those models do quite well. But the problem with those methods is that you can’t really figure out what they are doing and how sensitive they are to simply bad data.
If you take inaccurate parameters, the predictions might become lopsided.
For instance, doctors had predicted that the second wave wouldn't be devastating in India as about 50% of the people here already had antibodies for SARS-CoV-2. They even said we had almost reached herd immunity levels. But that 50% came from urban areas that were highly contaminated and they didn't take less contaminated zones in cities and the un-contaminated rural areas into account. In addition, there is also the possibility that a good percentage of the immune population loses immunity with time. Because the so-called Covid ‘supermodel’ (SUTRA model, which just had a data centric approach) commissioned by the Govt of India is fundamentally flawed. But still the government continued to rely on this model, than consult epidemiologists and public health experts. These numbers kept changing and their values relied on the number of infections being reported at various intervals. The SUTRA model’s omission of the importance of the behaviour of the virus; the fact that some people were bigger transmitters of the virus than others (say a gas cylinder supplier or a receptionist more than someone who worked from home); a lack of accounting for social or geographic heterogeneity and not stratifying the population by age as it didn’t account for contacts between different age groups also undermined its validity.
So we have seen the devastating consequences as everybody here dropped their guard.
Moreover, people all over the world largely ignored the mutating virus and the resultant highly infectious variants.
The models prepared in most places were severely flawed. This sometimes happens when we have a new virus and pandemic. We are still learning things and therefore, we cannot prepare accurate models using our limited knowledge.
One of the main reasons for the government's model not gauging an impending, exponential rise was that a constant indicating contact between people and populations went wrong. It was assumed it can at best go up to pre-lockdown value. However, it went well above that "due to new strains of virus".
Many epidemiological models extrapolate cases based on the existing number of cases, the behaviour of the virus and manner of spread.
According to experts 'no model, without external input from real-world data, could have predicted the second wave in India. However, the SUTRA model was problematic as it relied on too many parameters, and recalibrated those parameters whenever its predictions “broke down”. “The more parameters you have, the more you are in danger of ‘overfitting’. You can fit any curve over a short time window with 3 or 4 parameters. If you keep resetting those parameters, you can literally fit anything.
Models have always been important in science and continue to be used to test hypotheses and predict information. Often they are not accurate because the scientists may not have all the data. It is important that scientists test their models and be willing to improve them as new data comes to light.
Although modeling is a central component of modern science, scientific models at best are approximations of the objects and systems that they represent—they are not exact replicas. Thus, scientists constantly are working to improve and refine models.
This is a continuous learning process, and therefore, we cannot be accurate always.
So we keep getting these constant criticism from the world …. doesn’t matter, it is all part of the work.