none

Chinese’ 2020 Lunar New Year had been completely ruined. A virus-like SARS, MERS had brought a fast whirring economy to a halt, causing billions of dollars’ loss every day.  According to Chinese officials ’report on February 11, as of midnight on February 11, China ’s new coronavirus death toll rose to 1018. 43,099 cases were confirmed. (See the latest stats at https://www.worldometers.info/coronavirus/).  Models predicting the total number of being infected (including those not confirmed) will be around 250,000 in the next couple of days. To make it worse, we now know that many of the human sufferings during the process could have been prevented or avoided. Here is the summary of the main “pain points” from various Chinese media (including Tencent, Weibo)

The main complaint at the beginning of the Outbreak was “holding back the information about the infected cases for the sake of stability”, which might have made the containment harder.

The Chinese government is publishing detailed stats on the cases every day but the distrust in the government-issued stats has not only allowed disinformation, rumors to linger longer but also led to “inactivity” from people who could help. The reports of the Chinese Red Cross unfairly tilted distribution of donated materials such as face masks to government officials rather than doctors and hospitals, had many donations be sent to certain doctors, hospitals individually. (Local Red Cross under fire over China coronavirus donation mess), potentially causing inefficiency in matching the distribution to the most urgent needs.

One might wonder why the info on these key metrics (e.g. # of cases, severity of the cases, the needs for protection gear, test solution(Only 30% patients can get access to the RNA test kit, without which, the patients have no way to be admitted into hospitals), medical personnel, hospital beds, etc) cannot be managed by a centralized database. There are many individual reports on the ordeals for patients to get treatment due to the opaque nature of these key decision-making data. Many patients, especially, in Wuhan, not only suffered avoidable pain but also got infected by the coronavirus when they crowded the over-burdened hospitals in panic mode.

The most popular app, WeChat, in China has an inbuilt dashboard on the latest progress of the Outbreak (Follow the path: Payment/Medical Services, no payment necessary). However, it is not the whole picture view as there are many “ Likely Cases” (Defined as “ Patients that have a fever, being confirmed with CT exam with white spots on the lung, but NOT yet confirmed by RT-PCR /QIAamp viral RNA mini kit) not included. If Tencent or many talented programmers who are now forced to stay at home can enhance the App or creating a new app that will allow people to self-report their symptoms, log the progress of the symptoms, (with local community’s support to “verify” to minimize any chance of abuse), it should generate a big “structured” dataset (independent of any agenda) that data scientists, including AI, can crunch and provide more comprehensive, real-time decision making support insights to help the government to distribute resources more efficiently. A command center equipped with this live dataset should be able to allow:(Today’s command center makes decisions based on manually collected data from different levels. In most cases, it takes 2–3 days to confirm one coronavirus case. By the time the number reaches the Command Center or this “ WeChat dashboard”, it is already obsolete)

1. Prioritize patients to be sent to the nearest medical facilities based on their symptoms (vs. more than 90% of them go to the hospitals, waiting for hours, still be sent home. The doctors balanced the risk of further infection and the needs for immediate medical intervention). The majority of the non-urgent patients are much better off “self-quarantined” at home and follow medical instructions. A simple online request solution from the patients – Case Review – Approval – Notification- Setting up the Appointment (coordination) workflow will spare the pain in addition to the virus.
2. The SELF-reported data may also allow the medical community to spot potential mutation, turning point, treatment options (for certain demographics).
3. However, building this “self-reported” data set will not be easy as data accuracy will be one of the main concerns. Tech giants like Google had tried using big data to predict the onset of the epidemic. The now-defunct “Google Flu Trend”, aiming to predict flu outbreaks based on web searches of google users. It was successful (97% consistency with CDC data on flu cases) until 2011-2013: It predicted twice as many flu cases than Doctor’s visits. The inaccuracy of the dataset is partly due to the suppliers, many of whom were layman who is JUST searching “flu” related keywords, while they are not and will not be flu victims. Therefore, it will take the collaboration amongst doctors, public health experts, social workers, data scientists to provide context to different subgroups of the data set to apply different weighting to make the predictions more accurate.
4. If this dataset can be connected to a real-time tracking Supply Chain solution (Chinese commercial apps are very advanced in tracking the movements of goods) for Medical Resource Supplies, including the hospital beds, doctors/nurses, Ambulances (In Wuhan, there had been backlogs of days for 120 requests), the government can really live up to their claim that “ everything is under control”.

Virus outbreaks will never be eliminated but with today’s technology development, we got to finally established a Response System to minimize the loss of each time. It is not just a task for the Chinese Government but for every countries government as a virus does not respect country borders. Maybe an organization like Bill and Melinda Foundation can set up a team to develop an IT platform that will monitor and coordinate any outbreak. (The next outbreak? We’re not ready)