Alison Stribling has it he learned a lot about infectious diseases since he moved to the Covid-19 response at the Contra Costa County health department near San Francisco. One of his discoveries: the importance of fax machines in responding to the American pandemic.
Across the country, laboratories and health care providers are reporting new cases of Covid-19 to local health departments. At Contra Costa Health Services, officials use the data to initiate contact tracking or send additional help in certain cases, such as to a nursing home or an infected health worker.
On a typical Contra Costa day, only half of these reports come electronically; the rest, up to hundreds, pass through the fax line, creating a reading list of Sisyphus. “It can be a very long day, especially during floods,” says Stribling, a public health program specialist. “It’s the feeling of ‘I can never get over it.’
Now, Contra Costa’s early fax servers have some high-tech help. Just before Thanksgiving, the department launched software called Covid Fast Fax, developed in hasty collaboration with researchers at Stanford University. Dial the most urgent new faxes using machine learning algorithms. When Stribling and other fax colleagues got back to work after the holidays, they had a pending hundreds of faxes to read, but at least they knew where to start. “It was a good time,” Stribling says.
Like many other things about the response to the American pandemic, the project highlights the harshness of the country’s health care system. It’s also another example of creative minds combining it with hasty innovation, after car workers made facial shields or homemade hand sanitizer. In 2020, these projects can save lives. Contra Costa’s Stanford contributors have now published their code and methodology for other researchers or health departments to use.
Contra Costa got his assistant in artificial intelligence after Amit Kaushal, a Stanford professor and practicing physician working on integrating machine learning into health care, offered the department his skills this spring. Kaushal suggested collaborating on a grant he had received to work to curb the spread of the virus with a contact tracking app using Bluetooth signals. Officials got more excited when he launched the notion of an AI-enhanced fax line.
Contra Costa health officials struggled with more than just the volume of incoming case reports. Faxes appear as PDF files on a server, not on stacks of paper, after the 21st century. But it is difficult to detect and evaluate a Covid-19 case at a glance. Cases can be reported in several different ways, which are also used for other illnesses, often erased by hand without typing and sometimes reaching a lot of other messages or records. On a typical day, two public health specialists would be assigned to read and prioritize incoming faxes. “Very few faxes are the same and a lot of attention to detail and training is needed to know what you’re looking at,” says Stribling, who for a time led the team that handled incoming case data. “This can be difficult to do for eight hours or more.”
Kaushal and other Stanford researchers sought to domesticate the problem using machine learning, image-analyzing software, a technology most commonly targeted by medical researchers at tumors, not faxes.
To avoid the treatment of sensitive medical data, Kaushal recruited some other doctors to fill out disease information forms with patient data randomly generated in authentic doctor tracings. Fake forms were sent to a fax line to create authentic looking sample data. Postgraduate student Adam Lavertu used this data to form software to classify whether a page of an incoming fax contains a new Covid-19 case report or is something else, such as a medical record or a TB report.