Health Systems Turn to AI to Stitch Together Virtual and In-Person Care

If you've ever sat in an emergency room waiting to be seen while a nurse tried to juggle five other patients, you already understand the problem these partnerships are trying to solve. Healthcare has too few hands and too many people who need them. Over the past several weeks, a cluster of announcements from hospitals and telehealth companies suggests that artificial intelligence is being pressed into service not as a replacement for those hands, but as a way to stretch them further.

The pattern is worth paying attention to because it's not just one hospital experimenting. It's a coordinated shift across nursing, primary care, emergency medicine, oncology, and mental health, all leaning on AI to smooth out the handoffs between virtual and in-person care.

Take VirtuAlly, which is partnering with hospitals to deploy virtual nurses around the clock. Think of it like adding a second set of eyes and hands to a unit, ones that live on a screen rather than at the bedside, but that can still help with admissions, discharge paperwork, patient education and monitoring. Collette Health, which already serves more than 185 U.S. hospitals, announced last month it would combine its virtual observation and workforce management platform with VirtuAlly's services. The goal, according to Collette Health CEO Holly Miller, is straightforward: free up bedside nurses to focus on the patients in front of them.

"Hospitals need more than a monitor and a camera; they need experienced nursing judgment behind the technology," said VirtuAlly CEO Joseph Wechsler, describing what the partnership is meant to deliver.

The idea is spreading into emergency departments too. On Sept. 1, VirtuAlly announced it's expanding into ED arrival workflows with MUSC Health in Charleston, South Carolina. Under the new Virtual Provider in Triage program, a patient can be evaluated by a virtual nurse and provider almost as soon as they walk through the door, before a traditional treatment room even opens up. Testing and treatment can start early, and lower-acuity patients may be discharged straight from the front end of the department.

"When provider evaluation and virtual nursing can start shortly after arrival, we can move a patient's care forward before a traditional treatment space is available," said Dr. Marc Bartman of MUSC Health's Adult Emergency Department. "That creates a better arrival experience for patients while helping emergency departments use their people and physical resources more effectively."

Matching Patients to Trials, Doctors and Providers

Beyond the hospital floor, AI is also being used to connect patients with specialized care they might otherwise miss entirely. QC Healthcare announced on Sept. 2 that it's teaming up with Massive Bio, whose platform uses AI to match cancer patients with clinical trials, folding that capability into QC's newly launched My Adult Doctor virtual care service. That platform is staffed by board-certified physicians and now runs in all 50 states plus Puerto Rico and Guam, offering everything from urgent care to chronic disease management, behavioral health support, and GLP and peptide prescriptions.

The oncology piece addresses a real gap. Many cancer patients are treated in community clinics where oncologists simply don't have time to check every case against thousands of active trials. "This collaboration helps close that gap by surfacing relevant trial options for patients earlier in their care, wherever they happen to walk through the door," said Arturo Loaiza-Bonilla, Massive Bio's co-founder and chief medical AI officer.

QC Healthcare's CEO Eric Doherty was careful to frame the technology as a tool rather than a substitute for clinical judgment. "Artificial intelligence has the potential to fundamentally improve healthcare delivery, but success will depend on responsible implementation, clinical validation, strong governance, and keeping physicians at the center of decision-making," he said.

A similar philosophy is showing up in primary care. Atlantic Health, based in Morristown, New Jersey, has integrated K Health's PatientGPT tool directly into its patient portal, currently in beta. Patients can describe their symptoms, get personalized guidance, and book same-day, same-hour virtual primary care appointments, all without leaving the app. Atlantic Health CEO Saad Ehtisham argues this kind of AI actually makes care feel more personal, not less. "Technology should strengthen the connection between patients and clinicians, reduce barriers to care and help people get the right care at the right time," he said.

K Health's chief product officer Ran Shaul says the tool is built on the same foundation everywhere it's deployed: clinical grounding, clinician-led decision-making, and documented policies. A study at Cedars-Sinai, where Shaul discussed the technology last year with medical director Jonathan Weiner, found the AI adhered well to clinical protocols but that human doctors still outperformed it on nuance and complicated cases, a reminder that these systems are meant to support clinicians, not replace their judgment.

Meanwhile, Sword Health, which provides musculoskeletal and mental health care to employers, has filed to acquire OrangeDot, the parent company of Headspace, in an all-cash deal expected to close Sept. 14. The move brings Headspace's AI coaching platform, Ebb, under Sword's roof alongside its own clinically supervised tools: MindGuard, which screens messages for signs of distress, and Dawn, a chatbot that interacts directly with users. It's Sword's second major acquisition this year, following its $285 million purchase of Kaia Health's digital therapeutics business.

And in a sign of how differently people are now starting their healthcare journeys, Zocdoc is partnering with Google to let Gemini app users book in-network appointments directly through the chatbot. The company points to a 2026 survey finding that 65% of patients have used AI to figure out what kind of care they need. Providers already listed on Zocdoc automatically become bookable through Gemini, no extra integration required, as long as they have open slots.

Why It Matters

None of this technology fixes the underlying shortage of clinicians, and it shouldn't be mistaken for a cure to that problem. What it can do, if implemented carefully, is reduce the friction patients feel when trying to get care, whether that's a cancer patient searching for a trial, a parent typing symptoms into a portal at midnight, or someone waiting too long in an emergency room hallway. The companies quoted here are consistent on one point worth taking seriously: these tools work best when they support clinical judgment rather than substitute for it. Whether that principle holds as adoption scales up is the question patients, providers and regulators will need to keep watching.