How AI Can Predict Coronavirus like Epidemic Before it Outbreaks?

Spread the love

AI in healthcare is already developed enough to diagnosis various types of critical diseases, but in case of the epidemic it failed and not able to predict timely, that took the life of many people across the world and still spreading further became a health emergency.   

I’m talking about “Coronavirus Infection” – that started in mid-December in China and now transmitted to all major countries worldwide. This high contagious infection took the lives of more than tens of thousands of people and infected over 1 million people globally.

Also Read: How Exactly Coronavirus Attacks, Infects & Affects Body to Death

The question arises here, why artificial intelligence has not been used to detect the risk associated with this kind of disease or AI is unable to detect such epidemic with right predictions, so that medical experts can envisage the situation timely.

Artificial Intelligence Coronavirus Prediction

A Canadian based global health monitoring platform – BlueDot, reportedly notified its clients of the outbreak of coronavirus on Dec. 31.  But nobody has taken AI prediction seriously and now the situation became out of control in China.

BlueDot is the mastermind of Kamran Khan, who is an infectious disease physician and professor of Medicine and Public Health at the University of Toronto. Keep in mind that he was a frontline healthcare worker during the SARS outbreak.

How BluDot’s works in Epidemic Prediction?

BlueDot’s algorithm uses machine learning (ML) and natural language processing (NLP) technology to detect signs of potential disease outbreaks from the collected information that becomes a training data while developing such AI models.

Video: How BluseDot AI Predicted Coronavirus?

And such AI’s findings are reviewed and verified by human epidemiologists before sending a report to the company’s clients in government, industry and public health, as well as other public health officials, airlines and hospitals in the affected regions.

Data Used in AI Coronavirus Prediction

In the case of the coronavirus outbreak, the algorithm reportedly used airline ticketing information and pick news of such outbreaks like murmuring or forums or blogs or indications of some kind of unusual events going on to accurately predict the virus’ rapid spread from Wuhan, China, to Bangkok, Seoul, Taipei, Tokyo and other nations.  

Here, in coronavirus prediction, BlueDot uses an AI-driven algorithm that scours foreign-language news reports, animal and plant disease networks, and official proclamations to give its clients advance warning to avoid danger zones like Wuhan.

Another AI Predicts Coronavirus Could Kill 53 million and infect 2.5 billion

Yes, as per an article published on Forbes, AI predicts coronavirus could infect 2.5 billion and kill 53 million. But doctors said that it is not credible.

Actually, since the coronavirus infection transmission started more than 30,000 people infected and died around 600. But conditions of infection are changing, which in turn changes incredibly important factors that the AI isn’t aware of.

To predict this epidemic along with infection and death data, AI neural net using a recurrent neural network (RNN) model and ran the simulation ten million times. That output dictated the forecast for the following day. Once the following day’s output was published, added it to the training data, and re-ran ten million times, the results are shocking. 

ai coronavirus prediction
AI coronavirus unreliable prediction

From 50,000 infections and 1,000 deaths after a week to 208,000 infections and almost 4,400 deaths after two weeks, the numbers keep growing as each infected person infects others in turn. And in 30 days, the AI model says, two million could die and in just 15 more days, the death toll skyrockets enough to eliminate humans in millions. 

Artificial Intelligence in Medical Epidemiology Prediction

As per the report AI in medical epidemiology predicted dengue with more than 80 percent accuracy in Malaysia. AI in medical epidemiology predicted the outbreaks of dengue in Penang, Malaysia for 37 locations while the actual outbreak was 30, accounting more than 80% accuracy in prediction, making AI reliable in such epidemic prediction.

Artificial Intelligence in Medical Epidemiology

Though, scientists are developing ways to use AI to predict the spread of such contagious diseases before they happen. Though, the process is extremely complicated, successful implementation of predictive modeling could represent a major leap forward in the fight to rid the world of some of the most insidious infectious diseases.

How AI Can be Improved to Predict Coronavirus Like Epidemic?

However, as per multiple doctors and medical professionals, there is good news, the model doesn’t know every factor, as conditions and data fed into the neural network are changing and these conditions change, the results will also change.

However, in coronavirus like epidemic AI could predict the number of potential new cases by area and which types of populations will be at risk the most. This type of technology could be used to warn travelers so that vulnerable populations can wear proper medical masks while traveling or take other precautions to prevent such infections.  

Video: How AI Can Help to Control Coronavirus or Other Deadly Diseases?

Earlier researchers have built AI-based models that can predict outbreaks of the Zika virus in real time, which can inform how doctors respond to potential crises. AI could also be used to guide how public health officials distribute resources during a crisis. That will effectively work like AI stands to be a new first line of defense against such diseases.

Also Read: How AI is Used in Healthcare to Control the Coronavirus Disease

AI in healthcare is already playing a vital role in assisting with researching new drugs, tackling rare diseases, and detecting breast cancer. AI was even used to identify insects that spread Chagas, an incurable and potentially deadly disease that has infected an estimated 8 million people in Mexico and Central and South America.

And now AI increasing interest in using non-health data like social media posts helping health policymakers and drug companies understand the breadth of a health crisis. For instance, AI that can mine social media posts to track illegal opioid sales, and keep public health officials informed about these controlled substances spread.

The Uncertainty Factor While Relying on AI

One of the core strengths of AI is while identifying and limiting the effects of virus outbreaks is its incredibly insistent nature. The machine never tire, can sift through enormous amounts of healthcare data, and identify possible correlations and causations that humans can’t in a fast manner and if the amount of data is huge or very complex to analyze.  

While on the other hand, there are limitations of AI – the ability to both identify virus outbreaks and predict how they will spread. Let’s take the best-known example comes from the neighboring field of big data analytics.

At its launch, Google Flu Trends was heralded as a great leap forward in relation to identifying and estimating the spread of the flu — until it underestimated the 2013 flu season by a whopping 140 percent and was quietly put to rest.

Data Quality is Important for AI-based Predictions

Poor data quality was identified as one of the main reasons Google Flu Trends failed. Unreliable or faulty data can create confusion on AI-based prediction. In our increasingly interconnected world, tracking the movements of potentially infected individuals (by car, trains, buses, or planes) is just one vector surrounded by a lot of uncertainty.

But, BlueDot was able to correctly identify the coronavirus, in part due to its AI technology, illustrates that smart computer systems can be incredibly useful in helping us navigate these uncertainties.

And most importantly, it is not the same as AI being at a point where it precisely does so on its own – and that is the reason BlueDot employs only human experts to validate the AI’s findings.

Nevertheless, to ensure the accuracy of AI-based predictions for such an epidemic, a quality and reliable source of training data is necessary for supervised machine learning.

Also Read: Reasons Why AI and ML Projects Fail Due to Training Data Issues

Hospitals, medical centers and healthcare organizations need to share the labeled medical images of such infected people to AI developers, so that they utilize the same and help the medical science and AI engineers develop a reliable AI model.  

 CT Scan of a coronavirus infected patient in China
CT Scan of a coronavirus infected patient in China showing ground glass lesions in the lungs. Images Credit: Radiological Society of North America.

So that symptoms could be identified by the doctors and annotated to make it recognizable to computer vision through machine learning algorithms. In this case, the radiologist described Novel Coronavirus (2019-nCoV) Pneumonia through CT Imaging.

And when a huge amount of such CT scan images are manually annotated by experienced radiologists, it is used as a training data for machine learning AI models, that can in future detect such infections if similar symptoms are visible among the people. And as much as data used in the algorithm training, the accuracy of prediction by the model would be high.        

Leave a Reply

Your email address will not be published.