Technology can help diagnose, contain COVID-19 – within limits
Digital tools such as telehealth, remote patient monitoring, data analytics and even consumer-facing AI-based chatbots could play a key role in containing the outbreak of COVID-19 and help people who think they've been exposed to the novel coronavirus – but experts warn that such tools are not a cure-all.
For example, AI-based algorithms could be most helpful in providing information about patients who have already been diagnosed with COVID-19 or who are suspected of being infected.
Monitoring these patients remotely with clinical-grade sensors and collecting data on numerous physiological signals could improve clinical decision-making for providers, and the process can also help them learn more about the disease so they can better treat it.
The learning from the AI-based algorithms could then be combined with other information such as laboratory and imaging tests to create a composite mechanism that could help clinicians understand the disease better, and ultimately lead to better detection or prediction of the early signs of infection.
Kuldeep Singh Rajput, CEO of Biofourmis, a Boston-based specialist in digital therapeutics, told Healthcare IT News that there are clinical-grade wearables now available that can monitor patients with confirmed or suspected COVID-19 who are under quarantine, whether it's in a home or healthcare setting.
The wearable biosensors capture multiple physiology signs. And when combined with advanced analytics, Rajput says there is a "huge opportunity" to detect physiology changes indicative of clinical deterioration that require medical intervention.
"While there's still much the medical community does not know about COVID-19, there are known signs and symptoms such as fever, increased respiration and shortness of breath," he explained. "Physiological signals such as temperature, respiration rate and heart rate can be measured by wearables while a patient is in quarantine to detect worsening conditions and impending issues."
Dr. Saif Abed, director of cybersecurity advisory services at AbedGraham, explained that any platform that allows the general public to self-screen, virtually seek clinical guidance, be triaged or receive support during self-isolation/quarantine is going to be beneficial, not just in terms of diagnosis, but also in terms of containment as a preventative measure for public health.
"Containing something as transmissible as COVID-19 means we shouldn't be flocking to a family physician or emergency room at the slightest symptom, because that can exacerbate spread, so remote monitoring could be powerful," said Abed.
He noted that during a public health crisis, the more high-quality data that is available, the better organizations will be at forecasting and planning for multiple scenarios based on their likelihood.
"The right platform processing datasets at volume, speed and with sophistication could significantly impact our ability to take more precise measures for containment," he said.
Abed pointed to AI-based technologies like chatbots, saying they could be effective for triaging and guiding that general public for self-isolation.
"It will not be 100 percent accurate, as not every person will be classically symptomatic when they used a bot and might simply be an atypical case," he noted. "However, the goal at a public health level is to minimize spread so chatbots could be useful to guide human behavior and be an efficient resource when a health system is short on staff to take calls."
Rajput agreed chatbots can have some advantages. For example, they can engage people and provide education about their disease or ask them to take certain precautions.
"This type of automated engagement can support the interactions patients have with their clinicians, but its value does not come close to other AI-based tools such as AI-driven analytics combined with a clinical-grade wearable for remote monitoring," he noted.
Dr. Jay Anders, chief medical officer of Medicomp Systems, a provider of physician-driven point-of care solutions that fix EHRs, said with the outbreak of COVID-19, quick identification and accurate diagnosis of the condition is critical to help prevent the spread of the virus.
"Our clinical knowledge base team is constantly monitoring new and existing diagnoses, guidelines and terminologies to ensure our clinical data engine is up to date and accurate," Anders explained. "As new diseases or illnesses are identified, we enter the clinical guidelines into our database, along with any clinically-relevant terminologies, symptoms and information."
However, Anders noted that preventing the further spread of COVID-19 requires more than entering in the terminologies and diagnosis protocols into a clinical database.
"The quick identification of the virus is critical, and the best way to get clinical guidelines and terminologies in front of a physician is to provide them at the point of care," he said.
When treating a symptomatic patient, physicians should be presented with a clinically relevant diagnosis when documenting the chart note, because when armed with the most current information at the point of care, physicians can make accurate and timely diagnoses.
He said the greatest challenge of compiling and distributing clinical information is that existing EHRs are inundated with too much data that is neither structured nor accessible.
"When time is of the essence – which it currently is with the coronavirus outbreak – physicians do not have time to search through screens upon screens of data to find the information they need," Anders noted.
He explained that clinicians need data that is compiled, stored, and distributed in a structured, clinically relevant format, which not only allows for quicker identification of presenting symptoms, but also makes it easier to interpret and share with the Centers for Disease Control and Prevention and other physicians.
"To slow the spread of COVID-19, clinicians need tools that help them to quickly and accurately identify potentially infected patients, and efficiently and effectively share critical disease information," he said.