An ecosystem of innovation fosters tech-based solutions to COVID-19 challenges
Clinicians, engineers, and IT specialists work together at Penn on innovations that help doctors provide the best care for patients amid continued social distancing and coronavirus restrictions.
And while innovation in health care usually brings to mind new treatments and medicines, the efforts of clinicians, engineers, and IT specialists demonstrate the importance technological infrastructure for rapidly deployable, tech-based solutions so clinicians can provide the best care to patients amid social distancing and coronavirus restrictions.
The telemedicine revolution
In late March, telemedicine was key for allowing Penn Medicine clinicians to deliver care while avoiding potentially risky in-person interactions. Chief Medical Information Officer C. William Hanson III and his team helped set up the IT infrastructure for scaling up telemedicine capabilities and provided guidance to clinicians. Thanks to the quick pivot, Penn Medicine went from 300 telemedicine visits in February to more than 7,500 visits per day in a matter of weeks.
But far from seeing telemedicine as a temporary solution during the pandemic, Hanson has been a long-time advocate for this approach to health care. In his role as liaison between clinicians and the IT community in the past 10 years Hanson, helped establish remote ICU monitoring protocols and broadened opportunities for televisits with specialists. Now, with the pandemic removing many of the previous barriers to entry, be they technical, insurance-based, or simply a lack of familiarity, Hanson believes that telemedicine is here to stay.
“As the pandemic evolved, people were aware that telemedicine could help the health care system, as well as doctors and patients, during this crisis,” he says. “Now, there are definitely places where telemedicine makes good sense, and we will continue to use that as part of our way of handling a problem.” Other benefits include removing geographic barriers to entry for new patients, reduced appointment times, increased patient satisfaction, and reduced health care provider burnout.
Simple solutions for COVID-19 challenges
As the director of Penn’s Telestroke Program, neurologist Michael Mullen has experience diagnosing from a distance. This spring, telemedicine carts his group uses were repurposed in COVID ICUs. At the same time, Mullen and group wanted to expand their ability to assess stroke patients remotely, so he reached out to Brian Litt, faculty director of Penn Health Tech, to see how he could collaborate to create an analogous telemedicine station using readily available, cost-effective components.
GRASP lab researchers (from left) Bernd Pfrommer, Kenneth Chaney, and Caio Mucchiani assembling telemedicine cart prototypes in Levine hall earlier this spring. (Image courtesy of Kenneth Chaney and Bernd Pfrommer)
Rapid and simple solutions are at the heart of Penn’s ModLab, a subgroup of the GRASP lab focused on robots made of configurable individual components. As part of a COVID-19 rapid response initiative, engineers worked with Mullen to figure out a viable solution in record time. “The idea was to make it as simple and as fast as possible,” says graduate student Caio Mucchiani. “With robotics, usually you want to make things more sophisticated, however, given the situation, we needed to know how we could use off-the-shelf components to make something.”
Fellow graduate student Ken Chaney, postdoc Bernd Pfrommer, and Mucchiani came up with a plan that replicated the required specs of the existing telemedicine carts, including state-of-the-art cameras for detailed imaging as well as a reliable, easily rechargeable battery. The team then put together 10 telemedicine carts, assembling the prototypes with social distancing and masks at the GRASP lab in early April.
While changes to treatment approaches mean that these carts still require additional field testing, Mullen is still eager to expand the program, be it for diagnosing patients safely or educating medical students in an era of social distancing. “In the setting of COVID, when everything was getting crazy, it was remarkable to see the energy that GRASP brought to help,” adds Mullen. “Everyone was really busy, and it was amazing to see this group of people who wanted to use their expertise to help.”
Checking in on patients with COVID-19
As coronavirus cases in Philadelphia began to rise, a new platform was launched this spring to help COVID-19 patients recover safely at home. COVID Watch sends text messages twice a day to learn how they’re feeling, and, if they report feeling more short of breath, they are routed to speak with a nurse who assesses if they can remain safely at home, need a follow-up telemedicine visit, or need to go to the hospital. The platform and protocols for follow-up were developed by a team from the Center for Health Care Innovation with clinical design insight and support from Penn Medicine OnDemand and nurses in the Penn Center for Connected Care.
The goal, says COVID Watch Medical Director Anna Morgan is to provide patients a lifeline so they can get prompt support when needed. “We found that about 15% of COVID-19 patients enrolled in COVID Watch end up needing some kind of support,” says Morgan, “COVID watch allows us to be able to monitor and check in on 1,000 patients with a few nurses, so it’s a really efficient model.”
Penn Medicine's COVID watch program automatically checks in with patients at risk for complications of COVID-19 who are self-isolating at home. (Image credit: COVID watch/Way To Health)
David Asch, executive director of the Center for Health Care Innovation, says that automation was an essential component of obtaining this level of on-the-ground vigilance, which without automation would have required as many as 500 personnel hours every day. “COVID Watch is half-technology, half-human clinical support. It’s a way to connect people with care, and automation was a way to make it more scalable,” he says.
Initially developed by Asch and Kevin Volpp, Way to Health is the technology platform that powers COVID Watch and other bidirectional text messaging platforms. Researchers and clinicians were able to use the infrastructure already developed, and lessons learned from clinical trials done on existing platforms, to quickly develop a tool for COVID-19. “With COVID, it was something that we had to get up and running fast, and we needed a way to do analytics,” says David Do, a neurologist and clinical informatics manager who helped develop COVID Watch. “We needed to be able to look and, within the hour, know how many people had gone down each arm of the decision tree. Then, as soon as we made any change, we had to know what was happening an hour later.”
Way to Health’s Chief Operating Officer Mohan Balachandran says that having an existing platform as a starting point is beneficial, especially in terms of the speed at which the team can develop new iterations. “When a new program gets rolled out, there’s anywhere from five to 10 iterations. Doing that quickly is the challenge, and that’s where we really excel,” he says. Because of this, the COVID Watch team was able to identify metrics that were useful for elevating cases versus ones that increased call volumes without changing outcomes.
“We were on the cutting edge, innovating and doing things no one else could do because we controlled the platform and could turn on a dime,” says Asch. Other health care systems also adopted this strategy as a result.
The future of tech in health care
“One thing that clearly is coming out of COVID is that telemedicine is here to stay,” says Mullen. Whether it’s improving the patient experience, expanding specialist care to new patients, or improving the efficiency of medical treatments or clinical trials, there are lots of ways that technology can have a huge impact on the future of medicine.
RELATED
For the future of COVID Watch specifically, the automated texting platform will be used to evaluate patient health outcomes. This recently funded study will look at the data collected thus far to see if it helped patients stay out of the hospital and to determine if there are different levels of engagement and outcomes between different patient groups.
Asch adds that pairing automation with human support is relevant for many conditions in addition to COVID, such as managing diabetes or high blood pressure. And because Penn Medicine has already invested in the infrastructure required, it’s an area where they will continue to excel. “It’s a real win for COVID Watch. This was co-designed in real time with patients, doctors, nurses, and computer programmers to solve the needs we were learning about in real-time and in a very design-centric way,” says Asch. “Most places don’t have the infrastructure we have to be able to flexibly deploy this type of platform, but places like Penn Medicine can pull this off.”
“This all needs to be driven by clinical data, so there needs to be institutional support and a willingness to try,” says Balachandran about the importance of both IT infrastructure as well as an institutional environment that supports innovation and experimentation. “Because we have clinicians, technicians, and design folks on staff, whenever someone comes to the Center for Health Care Innovation with a problem, we have a complete ecosystem of folks available.”
Through his role as both a clinician and a software developer, Do also sees numerous opportunities for using technology to improve on existing models of patient care. “Usually in the health care sector, people think of technology as something they should look to the vendors for, but I think what we at Penn realize is that we need to push the envelope on what technology can do. What's important is a movement towards this idea that we could make an ecosystem of software tools that can be used for a particular job,” says Do.
This ecosystem is already well-established at Penn. For Mucchiani, who is eager to continue his research on robots that can help older adults, this will involve continued focus on finding simple solutions. “If you have a problem, instead of trying to make a solution that can be so complicated that it may not compensate the initial requisition of solving the problem, just focusing on the problem itself,” he says.
“In order to be a modern health care provider, we have to be able to do telemedicine,” says Hanson. “And we’re ahead of others in that we had infrastructure in place and converted it quickly.”
Understanding order to disorder at the atomic scale opens possibilities for next-generation electronic devices
A Penn team has developed insight into the chemical and geometric mechanisms underlying the synthesis of new 2D materials, paving the way for next-gen devices, biomedical applications, and cleaner, quicker energy conversion and storage.
Penn buildings achieve LEED certifications, showcasing commitment to sustainability
Three recent building projects at the University of Pennsylvania have earned LEED Platinum, Gold, and Silver certifications, underscoring Penn’s ongoing commitment to sustainable design and construction.
OpenAI, DeepSeek, and Google vary widely in identifying hate speech
Neil Fasching and Yphtach Lelkes of the Annenberg School for Communication have found dramatic differences in how large language models classify hate speech, with especially large variations for language about certain demographic groups, raising concerns about bias and disproportionate harm.
