Computer Vision Joins The Fight Against Coronavirus
The coronavirus epidemic broke out just before China’s New Year holiday — a period in which hundreds of millions of people visit their families, often in different provinces. Such large-scale population movements dramatically increase the risk of nationwide contagion. To help address this looming crisis, local tech companies have joined the fight to slow or halt the spread of the virus, quickly demonstrating the unique value of emerging technologies like computer vision. In the past two weeks alone, tech firms have:
- Implemented solutions for large-scale monitoring of body temperatures. Initially, airports and train stations had passengers wait in line, where staff manually took each person’s temperature. This posed a risk to personnel and passengers alike by creating crowded spaces that put people in close contact and increasing the chance of contagion. Computer vision startups SenseTime, Megvii, and DeepGlint released solutions to detect and monitor forehead temperatures. These solutions combine RGB and thermal cameras to uniquely identify masked faces, measure each person’s temperature, and alert staff when it identifies someone with a temperature above 37.5°C (99.5°F). Such systems have been deployed across the country and can monitor 300 people per minute, increasing efficiency and reducing the risk of contagion.
- Provided AI models to detect mask-wearing detection to developers for free. Since the coronavirus outbreak, the Chinese government has required people to wear masks at all times in public spaces like supermarkets, shopping malls, and office buildings. Last week, Baidu released a new AI model to detect when someone is not wearing a mask or isn’t wearing their mask correctly. Partners can quickly build solutions using an API or SDK on Baidu’s EasyDL platform or even download the AI model for free via Baidu’s PaddlePaddle project.
- Helped doctors analyze large volumes of CT scans. CT scans of lungs are a major resource to diagnose infection — but a single scan can generate hundreds of images. Given the large number of patients receiving CT scans, doctors in Hubei province — the center of the coronavirus outbreak — faced the prospect of reviewing millions of CT images per day. Yitu, a computer vision startup with a focus on medical imaging, deployed its system to analyze images from CT scans for coronavirus infection in four hospitals. This AI model can identify pulmonary ground-glass opacities and changes in density, range, and area and highlight and summarize symptoms to help doctors make diagnoses faster.
Tech companies have packaged their capabilities, including computer vision, to help society go through this pandemic. Enterprises should also know how to prepare themselves with technologies in general. Please take a look at our recent report, “Prepare Your Organization For A Pandemic,” or blog, Coronavirus’ Impact On Tech Markets Is Likely To Be Limited In Time And By Country — For Now, for more suggestions. And if you’d like to share some insights about your company’s strategy for emerging technologies, feel free to reach out to me directly or set up an inquiry by emailing firstname.lastname@example.org.