How to calculate health outcomes in covid-19 using Machine Learning?

Viral pandemicsare a severe threat. COVID-19 is not first,
and won’t end.

We are collecting and sharing about the virus. Hundreds of research teams
are combining their efforts to collect data and develop solutions around the
world.

Machine Learning helps in:

  • Identify who is at risk
  • Diagnose patients
  • Develop drugs faster
  • Finding existing drugs that can help
  • Predict the spread of the disease
  • Understand viruses better
  • Map where viruses come from
  • Predict next pandemic situations

Now, promote the research to fight this pandemic and prepare better for
next.

1. Identifying who is at risk 

Machine learning proved to be invaluable in predicting risk at many
spheres.

With medical risk, machine learning is potentially engaging in three key
roles:

  • Infection risk: the risk of a specific individual or group getting COVID-19.
  • Severity risk: there is a risk of a particular individual or group developing severe COVID-19 symptoms or complications that would require hospitalization or intensive care.
  • Outcome risk: the chance that a specific treatment will be ineffective for a particular individual or group, and how likely they are dying.

Machine learning helps in predicting all three risks. 

2. Diagnose Covid-19 Symptoms

When new pandemic hits and diagnosing individual is a bit challenging.
Testing on a large scale is difficult, and tests will become expensive at the
beginning. Any symptoms of COVID-19 expects to be concerned with contracted
disease, if the same symptoms are indicated with many other potentially, under
mild infections.

When it comes using machine learning, helps to diagnose 

covid-19, promising researching areas are:

  • By using face scans to identify symptoms such as a patient has a fever
  • By using wearable technology such as smartwatches look for identifying patterns in a patient’s resting heart rate
  • By using ML-powered chatbots to screen patients based on self-reported symptoms.

3. Speed up drug developments

In response to a new pandemic, it is critical to developing a vaccine
with a reliable diagnostic method, and a drug for treatment. The current
process involves a lot of trial and error that takes time with a couple of
months to isolate a viable vaccine patient. 

4. Identifying existing drug 

Companies spend a lot of time and money, getting new drugs approved.
They need to be as sure as possible that these drugs are unexpected with
harmful side-effects.

This process protects us but also slows down during a pandemic when we
need a faster response.

But there are thousands of drug candidates who do not have time to test
them to find the right choice.

Machine learning helps us prioritizing drug candidates much faster by
automatically:

  1. Building knowledge graphs 
  2. Predicting interactions between drugs and viral protein.

5. Predicting the spread of disease using social networks

Unfortunately, pandemics are caused by viruses that are difficult and
expensive to keep track.

One of the problems is that there might be a significant gap in time and
space between contracting the disease, developing the first symptoms, and
testing positive.

By interpreting the content of public interactions on social media, an
ML model assesses the likelihood of novel virus contamination. 

The model might not classify people on an individual level, but it can
use all of this data to estimatethe pandemic’s spread in real-time
and forecastthe stretch in the coming weeks.

6. Understanding viruses through proteins

A virus such as COVID-19 needs proteins, and how we get sick depends
entirely on how these proteins interact with our bodies. But interpreting them
is not an easy task.

The following use-cases provide a few examples of how ML improves our
understanding of viruses by analyzing their proteins.

7. Figure-out for attacking viruses

Epitopes are the clusters of amino acids found on the outside of
viruses. Antibodies bind to epitopes in which our immune system recognizes and
eliminate the virus. Finding and classifying epitopes is essential for
determining which part of a molecule to target and develop a vaccine.

Compared to traditional vaccines which contained as inactivated
pathogens, epitope-based vaccines are safest. They prevent diseases without risk
of potential for deadly side-effects. 

Locating a correct epitope can be a time-consuming and expensive process
with a new pandemic such as COVID-19, finding epitopes faster speeds up
developing effective vaccines.

8. Identifying hosts in the natural world

We are experiencing a zoonotic pandemic with a novel coronavirus as a
pandemic caused by an infectious disease that originates in different species
such as bats spreads to human beings. Viruses such as Ebola, HIV, or COVID-19
survive unnoticed in a natural world for a long time, waiting for a next
mutation and next opportunity to attack. These hide in animals called as
reservoir hosts that are unaffected by illness. 

Knowing who these reservoir hosts are vitalin fighting a
pandemic develops strategies to control the diseases spreading and prevent more
outbreaks. 

The classical approach in finding reservoir hostscan take
years of research, and there are still many orphan viruses have not been
matched to an animal host.

By looking at small subsets of species, speed up finding these pathogens
in the wild.

9. Predicting the risk of new pandemics

Accurately predicting a strain of Influenza will make a zoonotic leap
jumping from one species to another helps doctors and medical professionals
anticipate potential pandemics and prepare. 

For example, Influenza-A exists primarily in the avian population, but
it can jump to human hosts. Researchers are working on Influenza-A isolated
sixty-seven thousand nine hundred and forty protein sequences from the database.
They filtered these sequences so that datasets included only those influenza
strains with a complete series of eleven influenza proteins.

With ML, these researchers can identify potentially zoonotic strains of
Influenza with high accuracy levels. 

Conclude

Machine learning is an essential tool for fighting the current pandemic.
If we take this opportunity to collect data, pool our knowledge, and combine
our skills, we can save many lives both now and in the future.

    <!–

  • ONPASSIVE
  • ONPASSIVE
  • 26 December, 2020
  • –>

  • 334 Views
  • 59 1

<!–

–> Source

NOT to be Missed Hurry Up!

Leave a Reply