Fine design of epidemic prevention health code

COVID-19 is still spreading around the world, and prevention and control will become the norm. How to prevent and control it efficiently and finely, and minimize the impact on people, maybe we can make a fuss about health codes.

Now the function and significance of various health codes is actually mainly to provide data by big data statistics. But not fine enough. There are mainly the following problems:

  • How to ensure the authenticity of information
  • People in high-risk areas are not necessarily high-risk, so high-risk areas should not be divided, but high-risk people should be defined.
  • Only people who are in close contact with the confirmed patients are high-risk people. It is most important to inform these people in time and check them out.
  • The core of prevention and treatment of infectious diseases is early detection and early isolation. How to find it and isolate it as soon as possible is the most important thing.

Universal testing, screening out all patients, and then centralized treatment, this is the most ideal method. However, based on the actual situation, it is more difficult to realize this. Then set up more detection links, such as the temperature detection of residential buildings, which is this role. When the abnormality is detected, further nucleic acid detection will be conducted. If the diagnosis is confirmed, all close contacts will be screened again to find the car infected person. This should be done quickly and accurately. You can realize all accounts receivable, and you should check everything.

National epidemic prevention requires a national epidemic prevention system. Everyone and each unit play different roles in the whole system. These categories are probably needed in this system:

  • Brain center cloud computing platform:Such a large amount of data and responsible data storage certainly need a powerful data storage and technical platform, and China, which has experienced the test of Spring Festival travel rush and the Double Eleven, is not lacking in this ability.
  • All-round primary inspection:This requires our Zu Guoqiang’s strong organizational ability to do a good job of detection at every small, building subway and other personnel flow nodes. The main function is to find people with fever and abnormal health codes, and send them for inspection in time if abnormal conditions are found. Do early detection and early isolation.
  • Nucleic acid detection:Rapid and convenient detection can expand the detection range. In South Korea, PCR nucleic acid detection method was adopted on January 31st, and the diagnosis result can be obtained in 6 hours. The "drive-through" detection station can detect about 380 people every day. This practice can quickly deal with a large number of testers and effectively reduce the contact between medical staff and high-risk groups. It used to take 30 minutes for each person to test, but now it only takes 10 minutes to complete. This may be used for reference.

Face recognition real name login:Nowadays, whether it is the card opening of financial banks or the real-name authentication of the three major operators, remote face recognition verification has been realized, and the uniqueness of the national ID cards provides a data basis for face recognition. Relying on the camera of smart phone, face recognition login is completed. Thereby ensuring the correctness of the information. Face recognition login+mobile phone verification code. Don’t enter complicated information such as the place of entry, the place of household registration, and the place of residence.

Close contacts real-time storage:Fine epidemic prevention is to accurately find infected people and close contacts with confirmed patients. Now the way to find close contacts is too rude. One person is diagnosed, the whole building belongs to close contacts, and someone in a train is diagnosed and needs close contacts to check the relevant trains themselves. This greatly misses these potential threat groups. Or accidentally injure someone who is not really close to you. Using the Bluetooth technology of mobile phone, you can search for additional Bluetooth devices in real time and upload them for filing. Bluetooth can search for devices within 10-20 meters, and record the contact time of storage devices and the MAC address of devices. Within 10-20 meters, those who have been in contact for more than 60 seconds are close contacts in the true sense.

Real-time location information:The record of real-time location information of mobile phone can tell you where you have been in contact with close contacts when pushing information, and provide information judgment for self-infection rate through self-judgment of personal protective measures at that time.
Real-time push of abnormal information: when someone in close contact with an individual is diagnosed on the 14th, the mobile phone will automatically push the message. The message template involves the following information: time, place, number of contacts, contact duration, and the nearest nucleic acid detection point.

For example:

Hello; At 3:50 p.m. on April 6, 2020, a confirmed patient met with you for about 10 meters for 300 seconds near Chaoyangmen subway in Chaoyang District, Beijing.

Now the nearest nucleic acid detection point is the outpatient department of Chaoyang Hospital. It is recommended that you do a good job of protection and see a doctor immediately.

There are only two states of health code, no abnormality and abnormal. Face recognition login shows the status, which will be invalid after 1 minute. You need to log in again. All residential buildings only need to check the health code, and if there is any abnormal state, they will not pass, and urge them to do nucleic acid testing. If the testing is negative, the hospital data will be restored to no abnormal state after synchronization.

Exception definition:Within 14 days, if there is a confirmed case in the Bluetooth information record, it will be displayed as abnormal.

Eliminate anomalies:Go to the hospital for nucleic acid detection. If it is negative, the hospital data will be uploaded and the abnormal state will be automatically lifted. If it is positive, the system will automatically inform the user to contact the confirmed user, and then inform the contacted user to do nucleic acid testing.

At present, the health code is too rough, and the overall structure is centralized. From top to bottom, it lacks a dimension, that is, information communication between groups. Adding this dimension can realize timely tracking and timely push, and the whole process is more automated, reducing a lot of manual intervention.

To prepare for a protracted war, make full use of existing technical means, give full play to data intelligence, and fine epidemic prevention means to minimize the impact on personnel activities on the premise of accurate positioning. I didn’t mention the issue of data security in the whole article because we can’t give up eating because of choking at this stage. The whole design is still described by big lines. If you have any shortcomings, please brainstorm and comment.

Lao Zhang, everyone is a product manager columnist. AI product manager, focusing on natural language processing and image recognition. Now he is a partner of smart insurance startup company, hoping to communicate with entrepreneurs in the field of artificial intelligence.

This article was originally published by everyone as a product manager, and it is forbidden to reprint without permission.

The title map comes from Unsplash and is based on CC0 protocol.