How the edge is reshaping healthcare
Here's one problem the pandemic has underscored: A healthcare system that was not terribly efficient to begin with now seems stressed to the breaking point. Infectious diseases are on the rise. Fears about contracting COVID-19 have caused people to put off routine exams and treatments, making chronic conditions worse and contributing to an additional 20 percent rise in excess mortality. And to put the squeeze on hospitals even more, cancellation of elective procedures hits them directly in their P&L statement.
At the same time, people are living longer and requiring more end-of-life care. According to the World Health Organization, we're already facing a global shortage of 7 million health workers, a number that could reach 13 million by 2035.
If ever there was an industry ripe for further digital transformation, healthcare is it. And that transformation is happening at the edge.
Today, edge computing devices are being used to monitor patients remotely, automate the delivery of care, bring artificial intelligence to improve the speed and accuracy of diagnoses, track the vaccine supply chain, and much more.
Please read: Healthcare Rx: How technology and IoT can help fix a broken system
"Healthcare systems are dealing with an increasing volume of digital data every day," notes Sid Shah, program manager for Frost & Sullivan's transformational health practice. "But relying on the cloud or on-premises servers means they're limited by data transmission speeds, bandwidth, issues around privacy and security, and cost. The ability to process data locally on the edge addresses all of these challenges."
That's because edge computing puts data, analytics, and processing power where it's most needed: in the ambulance, in the hospital, in the operating room, in our homes, and ultimately inside our own bodies.
How AI powers the edge
You can't talk about edge computing and healthcare without also talking about AI. It's not enough to collect data from patients; caregivers also need to analyze it and respond in real time. Increasingly, that process is being performed by machines at the edge.
The average hospital bed has between 10 and 15 connected devices, according to a report by IoT security firm Zingbox. A 2020 survey by Optum, the technology division of UnitedHealth Group, found that 40 percent of healthcare executives plan to deploy AI to analyze data collected by these devices.
Please read: How data and AI will shape the post-pandemic future
In fact, you'll find AI-powered edge computing solutions in nearly every scenario where healthcare is being delivered.
In the ambulance
Today, ambulances are primarily used to ferry patients to hospitals as quickly as possible. Within a few years, they'll transform into mobile edge computing platforms that help save more lives.
In Barcelona, for example, first responders inside vehicles have used tablet PCs to capture high-definition video and vital signs of trauma victims and then transmit the data over a 5G connection to emergency room personnel.
Please read: Managing emergencies in a not-quite-connected world
Emergency medical technicians can collaborate with doctors on how to stabilize the patient, while ER personnel can prep the room to handle their specific care needs, says Weisong Shi, a professor of computer science at Wayne State University and a researcher in the field of edge computing and connected health.
"By enabling edge computing inside emergency vehicles, EMTs can transmit crucial data to the hospital in real time, arming emergency department teams with the knowledge they need to save lives," says Shi.
In the hospital
Within hospitals, edge computing and AI are enabling faster, more accurate diagnoses and automating the delivery of medicine.
Instead of wheeling patients to a separate facility and waiting hours for scans to be processed, caregivers at UCLA Health, Massachusetts General Hospital, and King's College Hospital in London are starting to deploy edge devices like Hyperfine's portable MRI machine, which can capture brain scans at the patient's bedside. Machine learning algorithms speed image processing and identify anomalies, allowing radiologists to analyze the scans in real time.
Diabetes patients today can rely on an automated insulin delivery system that uses artificial pancreas sensors inserted under the skin to monitor blood sugar levels and transmit the data to an insulin pump and a handheld device. Algorithms inside the pump device predict where blood sugar levels are likely to be, then direct an external pump to deliver the appropriate amount of insulin.
Please read: Podcast: How intelligence at the healthcare edge saves lives
Processing glucose data in the cloud or an on-premises data center and sending instructions back to a wearable device is impractical at best, explains David C. Klonoff, MD, medical director of the Diabetes Research Institute at Mills-Peninsula Medical Center and an early researcher into diabetes technology.
"Many of these medical devices must respond immediately to sensor data, within milliseconds," says Klonoff. "They can't function properly if they need to wait for data to be sent to a remote location, analyzed, and then sent back. The latency for controlling an artificial organ like this is not compatible with analysis of data on the cloud. Edge computing is needed."
Eventually, he adds, we'll have tiny sensors like this implanted in our bodies that read our vitals, send alerts to our doctors, and trigger actions like the delivery of medicine or electrical stimulation.
In the operating room
Another area where edge computing and AI are transforming healthcare is the operating room. Frost & Sullivan predicts that the market for AI-assisted surgery will more than triple by 2024, exceeding $225 million.
For example, Zimmer Biomet is currently piloting an "intelligently analytical operating room," powered by Chooch Intelligence Technologies' computer vision platform.
During surgeries, nurses are required to log every action, from the patient being wheeled into the room to final cleanup. This process can involve pushing dozens of buttons on a touchscreen over the course of a procedure.
Using cameras and edge computing devices, the AI software automatically records and categorizes each action inside the OR. Data from hundreds of similar surgeries can then be aggregated and analyzed, leading to more efficient processes and better patient care, notes Chooch CEO Emrah Gultekin.
"Nurses can focus on the patient instead of the software," he says. "And if a process that's supposed to take 45 seconds suddenly takes 53 seconds, the system can alert the doctors in real time that something may be wrong."
In our homes
A primary goal of digital transformation in healthcare is to deliver treatments directly to patients where they live, increasing the availability of care while slashing costs.
One of the few bright spots to emerge from the pandemic is the rapid acceleration of telemedicine. According to the American Medical Association and Wellness Council of America, almost 75 percent of all doctor, urgent care, and ER visits could be handled safely and effectively by telemedicine.
As Zoom sessions replace in-office exams, telehealth visits will continue to grow at nearly 40 percent per year through 2025, according to Frost & Sullivan. The movement toward in-home care is also being enabled by a raft of edge devices, from digital stethoscopes and sleep trackers to motion sensors that use Wi-Fi signals to determine if you've fallen and can't get up.
Please read: Telehealth: How virtual can medicine get?
Using edge sensors to monitor patients in their homes can alert caregivers to changes in a patient's condition and allow them to modify treatments as needed, notes Wayne State University's Shi.
"Wearable sensors will help track a patient's health and send abnormal data to the doctor," Shi says. "Rehospitalization rates can be gravely cut down by enabling providers to keep tabs on their patients."
But the edge's greatest impact may be enabling telehealth services, remote patient monitoring, or drone delivery of lab samples and medical supplies to serve people who currently have little or no access to quality care.
"Edge computing enables doctors to deliver care to remote areas where connectivity may not be great and skilled healthcare personnel may not be available," says Frost & Sullivan's Shah. "It offers the ability to process data locally and provide clinical decision support to even low-skilled healthcare workers.
Cheaper, more private, more secure
Putting healthcare on the edge offers other ancillary benefits. For example, radiology scans generate enormous volumes of data, which can consume large amounts of expensive bandwidth and cloud storage. Storing this information locally is not only cheaper, but also helps protect the privacy of very sensitive and highly regulated patient data.
Isolating edge devices from the larger hospital network can also make the data less vulnerable to external attacks—but only to a point, notes Mike Meikle, global enterprise architect for a leading security firm and longtime healthcare security consultant.
"Keeping the data local can help with HIPAA compliance and privacy laws," says Meikle. "But at some point, they'll still need to move the data from the device to the doctor. And as we can see from the news cycle, healthcare orgs and IoT devices don't have a great track record when it comes to cybersecurity."
The robo doc is not in
The ultimate goal of digital transformation in healthcare is not to install a robot in place of your physician, explains Rich Bird, marketing manager for worldwide healthcare and life sciences at Hewlett Packard Enterprise.
"We're not replacing doctors or care teams with this technology," explains Bird. "We're helping them make quicker and better-informed clinical decisions by generating insights from providing data that helps improve the outcomes for their patients and reduce the cost of the care they deliver."
Over time, as medical services move even farther away from hospitals and doctors' offices, we'll reap the benefits of superior care, lower costs, and longer, healthier lives.
"In 10 or 20 years, healthcare will be very different," Shah says. "Hospital operations will be smoother, physicians will have access to actual insights to help them make better clinical decisions, and patient outcomes will improve. Patients themselves will be empowered to make better health decisions, which will lead to lesser complications later in life."
That sounds like just what the doctor ordered.
The latency for controlling an artificial organ ... is not compatible with analysis of data on the cloud. Edge computing is needed.
This article/content was written by the individual writer identified and does not necessarily reflect the view of Hewlett Packard Enterprise Company.