Since the first report of coronavirus (COVID-19) in Wuhan, China, it has spread to at least 100 other countries. As China initiated its response to the virus, it leaned on its strong technology sector and specifically artificial intelligence (AI), data science, and technology to track and fight the pandemic. In a global pandemic such as COVID-19, Data and AI have become critical to helping societies effectively deal with the outbreak.  As a result, we at IBM are integrally involved with clinicians, academics, and government entities around the world to activate technology as the virus continues to spread to many other countries.

Data, Analytics in combination with AI might be powerful tools to contain the spread of the COVID-19 epidemic and minimize its resulting fatalities. There are tremendous learnings from Taiwan, Hong Kong or Singapore (countries that were hit by SARS in 2003 and learned from it) on how to leverage the power of Data to manage and fight Coronavirus disease. They all managed to collect and correlate data from multiple sources, analyzed and visualized it and drove predictions on what might happen next. The key has been to integrate the national health insurance database with immigration and customs databases to create big data for analytics and AI. This in return allowed them case identification by generating real-time alerts during clinical visits based on travel history and clinical symptoms of patients. 

The most basic challenges Healthcare is facing currently

1. Hospitals and Medical Call centers cannot cope with all the questions from citizens
2. Hospitals are looking for ways to enable Triage and handover to Telehealth
3. Hospitals do not have insights in regards to who, where and how many people have questions

HealthCare Assistant to reduce public exposure by allowing citizens to run a self-service coronavirus assessment tool. 

AI to help better communicate and streamline messaging.

As nations impose national lockdowns to restrict the spread of Covid-19, causing a major influx of increasing calls to hospital and Government emergency lines, decreasing the call loads and providing accurate information becomes imperatives. The greatest challenge in producing the citizen assistant is ever-changing information about the disease daily and hourly as the virus spreads across length and breadth of Nations. Limited resources of clinical experts and the rush at the stage of epidemic spread. The scheme of the Citizen Assitant had to be changed several times.

The whole platform has to be configured to rapidly deploy an online COVID-19 triaging workflow. I would divide the questionnaire into two parts. Starting with the Base part related to disease more accurately answered from trusted sources like WHO and CDC. Second, customized and localized more answered from the local health authorities like in Singapore from the Ministry of Health, Australia from the Department of Health, India from the Ministry of social welfare, etc.

Base:  

• “What are the symptoms of COVID-19?”
• "What are the global travel restrictions?"
• "How do I protect myself from COVID-19?"
• And additional common questions related to COVID-19.

Customized and localized:  

• Where can I get tested for COVID-19 in (locality)?
• Is the local high school or polytechnics open?
• Is Singapore General Hospital accepting patients open?

In addition to the above, Citizen Assistant should be capable of closing the loop with the caregivers and health providers.

• Users have instant access to an online questionnaire that checks for COVID-19 related symptoms. 
• The questionnaire is automatically screened and triaged and creates a worklist for care coordinators.  
•  Care coordinators can then set up a teleconsultation with symptomatic patients. 
• If patients are advised self-quarantine, it automatically sends out alerts based on a care plan to regularly monitor their vitals. 
•  For patients who would require COVID-19 testing, the system can initiate referrals to clinical facilities.  

AI to identify, track and forecast outbreaks

The better we can track the virus, the better we can fight it. By analyzing the text in Citizen Assistant logs and mining the text exchanged, healthcare providers have a better chance of predicting where the outcome is likely to happen and take mitigating action. The key to this prediction is Data and AI. It uses natural language processing and machine learning to cull data from hundreds of chats citizens had with the citizen assistant combined that with statements from official public health organizations, digital media, global airline ticketing data, livestock health reports, and population demographics. It’s able to rapidly process tons of information every 15 minutes, 24 hours a day. 

Beyond just detecting infectious disease threats as early as possible, AI solution's goal is to understand how diseases might disperse to different parts of the world, and then determine the potential consequences of it spreading.

Taiwan officials from the beginning of the viral outbreak “did a very detailed mapping of who got it from whom,” and were able to stop a lot of transmissions early 

Notably, officials integrated Taiwan’s national health insurance database with its immigration and customs database. This enabled the government to track the 14-day travel histories and symptoms of its citizens, nearly all of whom have an identifying national health insurance (NHI) card. All hospitals, clinics, and pharmacies were given access to this information for each patient. 

Taiwan restricted entry for foreign travelers from the most affected regions, and for those allowed entry, officials tracked them with mobile technologies. Foreign visitors are asked to scan a QR code that takes them to an online health declaration form where they provide contact information and symptoms. People placed under quarantine are given government-issued mobile phones and monitored with calls and visits. 

How IBM is helping?

Watson Assistant for Citizens empowers Healthcare, Government and Educational organizations to build and deploy an AI-powered, compliant, conversational healthcare experience at scale. The Citizen Assistant combines built-in medical intelligence and intents with natural language capabilities allowing healthcare organizations such as Providers, Payers, Pharma, Telehealth to give people access to trusted and relevant healthcare services and information. The Citizen Assistant can be customized to suit Emergency Response and Government Service Delivery Scenarios.

We are helping front-line healthcare workers detect and monitor the disease efficiently. Imaging departments in healthcare facilities are being taxed with the increased workload created by the virus. The AI visual recognition and analysis tools like Watson Visual Recognition improves CT diagnosis speed.

Using Watson Assistant, IBM has designed a virtual assistant pre-loaded to understand and respond to common questions about COVID-19.