Application of AI, Bigdata in monitoring, detecting and preventing diseases caused by Corona virus

Tram Ho

With the current outbreak, related to a coronavirus originating from Wuhan and so far has infected nearly 16,839 people, 362 died and appeared in 30 countries and territories (Updated data. continuous), this is a lateral epidemic that the World Health Organization declared a state of emergency on January 31, 2020.
Because of the danger and the impact of this virus, many people are confused, not only in the focus of China, their Vietnamese neighbors or Americans but also globally. In this article, update some useful information that IT application in monitoring, detecting and preventing disease caused by Corona virus, (not the treatment or prevention of viruses).

Visualize disease situation data

The first is to use data to display continuous, immediate and intuitive disease data (Data Visualization). The data used to inform the tool is drawn from various sources. Information is mainly collected from organizations such as WHO and Centers for Disease Control and Prevention (CDC), China CDC, ECDC. For national internal, data is collected from the appropriate authorities and health departments.
This data shows the circles representing confirmed infested areas by country / province / state. You can click on each one to get some infections, deaths and recovery. Smaller tables surround the map with additional data such as a list of areas organized from most cases at least, charts of infection over time and a list of all countries where coronavirus has attacked. work. Example Click on California, and you will see two confirmed cases. The number of deaths and recovery was also monitored.
The dashboard is designed to give the public an understanding of the disease situation as it happens, with transparent data sources, CSS CSSE said in a notice posted on its website. Of course, the actual number of cases is unknown, but the panel at least provides reliable data for reported cases and can indicate trends and hot spots for coronaviruses.
Currently there are many sources to be able to follow the online display of this disease situation map:

Actively build a coronavirus monitoring dashboard (2019-nCoV) with Data Analysis

The outbreak of the epidemic has also created countless unverified or fake news shared on social media that only makes the situation worse by causing bewilderment for the public. Instead of sitting idle, LEAD CEO Dr. Lau Cher Han decided to take the proactive path to forming a team of volunteers, including IT experts, data scientists, and care professionals. health and public interest to build a web application tool that helps the public track the development of the deadly coronavirus 2019-nCov in real time.
The domain was registered the day before the Chinese New Year celebration. Dr. Lau later formed a public group on Telegram, inviting volunteers on Facebook to join and help build the tool – following the Hackathon method. On the first day alone, more than 100 volunteers from Malaysia, Australia, Japan, the Philippines, Singapore, Taiwan, the United States and other countries participated in the call for weapons to work on
In just one day, hundreds of volunteers wrote their ideas and plans on the Trello board.
Using the OSEMN framework in data science, the project kicked off with data collection, cleaning, and exploration followed by modeling and having the data presented visually on the site. Scrap sites are built in Python and BeautifulSoup, so news from accredited and authoritative websites, including websites presenting data on coronavirus spread.
The goal of’s usability is to be a news aggregator that summarizes news from recognized and authoritative outputs, using data science methods, such as NLP (processing natural language) to analyze the content and identify meaningful topics. Using AWS as a platform for storing scraps, web APIs and websites, the first section of was built with Vue.js and the backend with Node.Js and ExpressJS – with the help of Some volunteers are Full Stack developers and UI / UX designers. A pool of data has been deployed to render raw data and using MySQL, raw data is converted into structured data for presentation.
By the end of the two days, the site was up and over 11,000 people have used the platform to track coronavirus development around the world and in their region. home page is where you can find reliable articles. Each article listed here is also filtered by a group of volunteers, data scientists and health professionals. You can also filter news, analysis and refinement centers by country and state, to learn more about developments in your area.
A key feature of is the real-time analysis dashboard, where it has important data and their visual representation including the total number of confirmed cases, total number of deaths, flare-time. broadcast and countries affected by the epidemic. Analytical dashboard visualizes development on Coronavirus 2019-nCov in the near future. Data for the dashboard is mainly taken from CDC, JHU, Tencent and more. Although other tracking sites exist, CoronaTracker takes a few more steps to track coronavirus disruption over time.

What next in the plan?

At the time of this writing, more refinements are being made to, to remove more data from various local websites, such as Chinese websites and to add More features go to the site, like the location map of the current case. A mobile application for the public to receive real-time push notifications on updates is also in progress.
Want to contribute can join the coronatracker group on telegram : CoronaTracker Telegram Group
You can access the removed data to make your own analysis or prediction here: CoronaTracker Analytics on Github

Chinese technology companies use big data to analyze trends in internal migration

Baidu Maps, for example, can access and track migration trends, based on handheld devices and computers originating from the center where heavy infections were found, especially in Wuhan in Ho Chi Minh City. North of China, In addition, Baidu Maps provides real-time notifications on travel tips and road closures, as well as allows users in more than 200 cities to search for nearby fever clinics.
Similarly, mapping company AutoNavi also allows nearby fever clinics to easily check for and provide extensive information about the virus. It’s also easy to find a special emphasis on viruses on Alipay’s homepage, Alibaba’s popular mobile payment solution, which provides users with real-time data about viruses, the gateway to delivery services. food and shopping, among other mobile services that help people weather hard times.
In other cases of technological perspectives in the coronavirus war, Baidu recently announced that its smart outgoing call platform was freely opened to authorities at all levels, health committee agencies, community and disease prevention center starting from Monday until the end of the boom. The call platform has a feature that filters out the flow of migrants and local residents and gives notifications to designated groups of people and is said to be much more efficient than people’s phone calls.
Using AI (artificial intelligence) is a useful tool in global outbreaks.
Artificial intelligence will not prevent new coronaviruses or replace the role of expert epidemiologists. But for the first time in a global outbreak, it is becoming a useful tool in crisis monitoring and response efforts, according to health data experts.
In previous outbreaks, AI provided limited value, because it lacked the data needed to provide quick updates. But in recent days, millions of coronavirus posts on social media and news sites are allowing algorithms to create near real-time information for public health officials to track the spread. its.
The field has grown tremendously, says John Brownstein, a computational epidemiologist at Boston Children’s Hospital who runs a public health monitoring website called that uses AI to analyze data from Government reports, social media, other news sites.
During the SARS period, there was not a huge amount of information coming from China, he said, referring to a 2003 coronavirus outbreak from China, infecting more than 8,000 people and killing nearly 800. At the moment, we’re constantly exploring news and social media.
Brownstein emphasized that his AI is not intended to replace public health leaders’ information collection work, but rather to complement their efforts by compiling and filtering information to help them put in place. decision making in fast changing situations.
We use machine learning to scrape all information, categorize information, tag it and filter it – and then that information is passed on to our colleagues at WHO who are reviewing this information. all day and evaluate, Mr. Brown Brown said. There is still a challenge in parsing whether some of that information makes sense.
These AI monitoring tools have been available in public health for more than a decade, but recent advances in machine learning, combined with greater data availability, are making them powerful. much more. They also allow use beyond basic surveillance, to help officials more accurately predict how quickly and widely spread, and what type of people are most likely to be affected.
Don Woodlock, vice president of InterSystems, a global electronic health provider, said machine learning is good at identifying patterns in data, such as the risk factors that can determine zip codes. or the cohort of people involved in the virus. The record is helping suppliers in China analyze data on coronavirus patients.
When different treatments are tested, he added, we can also use machine learning to determine what might work with the virus.
It’s still too early to break those types of analysis, but AI tools can help accelerate that research once more data is available. The true impact of AI in dealing with the coronavirus outbreak probably won’t be known for several years.
Brownstein said efforts to harness the power of AI to predict the course of the disease – and the scale of the impact – are taking place at breakneck speed. Groups across the country are developing domestic and international spread patterns (of coronavirus), he said, adding that is partnering with a Boston-based startup called Buoy Health to build a symptom testing tool to assess symptoms of coronavirus. Distinguished from seasonal flu.
That promises to be a major challenge for public health officials in the coming months, as they work to allocate resources to accommodate viruses and manage a range of possible cases to the clinics. emergency. The more we focus on intervention efforts, identify cases as soon as possible and isolate those cases, the more we have a chance to limit the global impact of this virus, ” said Brown Brownstein. .

How does BlueDot predict coronavirus using artificial intelligence (AI)?

In the time of the virus outbreak that China and other countries are facing, time is of the essence. Warning as soon as possible, the opportunity to stop the spread as possible.
One problem, however, is that governments are sometimes wary of sharing information. That was the case in 2002 and 2003, when Chinese authorities were accused of covering up the SARS epidemic, which eventually claimed more than 740 lives around the world.
But even if Beijing provides less information, the world now has better information tools its way than it did 17 years ago. One is provided by Bluedot, a Toronto-based startup that has an AI-driven health monitoring platform analyzing billions of data points. Launched in 2014, the joint venture warned its customers about the outbreak on December 31, ahead of announcements from the World Health Organization and the U.S. Centers for Disease Control and Prevention.
The company says it uses big data analytics to track and predict the spread of the world’s most dangerous infectious diseases. Last August, they announced an investment round that brought a total funding of about $ 10 million.
Bluedot uses natural language processing techniques and machine learning to screen through global news reports, aviation data and animal disease reports, as described by Wired. The epidemiologist reviews the results automatically, and if everything checks out, the company sends notifications to its clients in the public and private sectors.
BlueDot tries to track and move information faster than the disease can move. It predicted exactly where outside of mainland China, the Wuhan virus will land in Bangkok, Bangkok, Seoul, Taipei, and Tokyo, right after its first appearance.
Company founder Kamran Khan told the Canadian press that on the one hand, the world is changing fast, where disease is emerging and spreading faster. On the other hand, we happen to have increasing access to data, we can use BIG to gain knowledge and spread them faster than infectious diseases.

AI is the future of health care

It’s clear that AI and Machine Learning techniques will undoubtedly be the future of the healthcare industry and can disrupt the industry forever. According to Frost & Sullivan, AI systems are expected to be a $ 6 billion industry by 2021. A recent McKinsey study predicted health care as one of the top 5 industries with over 50 AI-related use cases, and more than $ 1 billion in startup equity.
AI in healthcare will fundamentally affect the three main aspects of health care – Patients, Doctors and Administration / Activities. While AI tools and bots will be deployed at every level of a patient’s medical journey, it’s the overall impact that it will make that will really disrupt the industry. The AI ​​will continuously gather all a patient’s past records, along with insights, use that data to diagnose, treat, and ultimately maintain, health.
Summary: As the world changes rapidly, these diseases are emerging and spreading at a rapid rate. However, with various AI tools and software, increased data access can be used well. Significant increases in data can be used to create important insights and in turn act upon them thus spreading the news earlier and faster than the disease can spread itself.
Share the news now

Source : Techtalk