Why We Built Implansense: One EP Clinic's Alert Overload

The problem was visible before I had the vocabulary for it.

In the early months of 2024, I spent a significant amount of time embedded with device clinic teams at a pair of Philadelphia-area cardiology programs — observing workflows, asking questions, understanding what the day actually looked like for the nurses who managed post-implant remote monitoring. What I saw in those clinics wasn't a technology failure. The devices were transmitting reliably. The vendor monitoring services were doing what they were designed to do. The data was arriving.

The problem was that by the time the data arrived in the clinic queue, every transmission looked about equally urgent. A routine quarterly ICD check sat next to an alert from a patient who had received an ICD shock overnight. A pacemaker battery status report sat next to a transmission showing a significant AT/AF burden increase in a patient on subtherapeutic anticoagulation. The nurses I spoke with described the same coping mechanism, independently of each other: work through the list in order, trust your instincts when something looks off, hope you haven't missed anything by noon.

What the vendor apps don't solve

It's important to be precise about what the major device manufacturer monitoring services — Medtronic CareLink, Abbott Merlin.net, Boston Scientific LATITUDE, Biotronik Home Monitoring — actually do. They do their jobs very well. They reliably capture device telemetry and surface manufacturer-defined alert thresholds. The clinical engineering teams that built those platforms have deep device-specific knowledge. The alert logic in those systems reflects genuine expertise.

But those platforms were built to be the data source — the authoritative record of device performance for each manufacturer's devices. They were not built to be the clinical workflow layer for a device clinic managing 2,000 patients across four different manufacturer ecosystems. They don't know which vendor's transmission is the most urgent event in the clinic's queue this morning. They don't compare urgency across their own alerts and another vendor's alerts. They don't have visibility into the patient's recent clinical history, hospitalization status, or whether the nurse manager is operating with two-thirds of her usual staff today because two people called in sick.

That cross-vendor, cross-context clinical prioritization layer is what didn't exist. And what I kept hearing from device clinic nurses, device clinic medical directors, and EP physicians was some version of the same thing: "We need something that tells us what matters first. The vendor apps tell us everything equally."

The moment the problem became concrete

The clearest articulation of the problem came from a device clinic nurse manager at a regional cardiac center outside Philadelphia. Her program tracked approximately 1,400 active devices — not an unusually large panel, by the standards of hospital-based programs. On a typical day, her team received somewhere between 50 and 120 inbound transmissions across the major monitoring platforms.

"On a slow day, we can work through everything by three o'clock," she told me. "On a busy day, we're still reviewing transmissions at six. And the ones we got to at six — I can't tell you whether one of them should have been handled at nine."

That phrase — "I can't tell you whether one of them should have been handled at nine" — is the founding insight of Implansense. It describes a structural problem, not an effort problem. The people in these clinics are skilled, experienced, and conscientious. They're working in a system that doesn't give them the information they need to prioritize effectively at the start of the day. The solution isn't asking them to work faster. It's building the layer that does the pre-sort before the clinical judgment happens.

Why we started with triage, not EHR integration

There's a legitimate question about why Implansense focused first on alert triage logic rather than starting from the EHR integration problem. The EHR gap is real — remote monitoring data living outside the clinical record creates documentation friction, and closing it is clearly valuable. We're building toward that.

But the EHR integration problem, while difficult, is primarily a technical interoperability challenge. HL7 FHIR standards exist. SMART on FHIR frameworks exist. The path from device telemetry to structured EHR data is navigable.

The triage problem is different in kind. It requires understanding the clinical weight of device parameters across multiple device types, normalizing data from multiple vendor systems that were not designed to talk to each other, and building scoring logic that respects the clinical variability between device types and patient populations. That's the harder problem, and it's the one that has the most direct impact on patient safety when it isn't solved. Alert fatigue in device clinics is a daily-occurrence problem, not an edge case. We built for the daily occurrence first.

What Implansense is — and what it isn't

Implansense is a clinical analytics layer that sits between the device manufacturer remote monitoring services and the EP clinic workflow. It ingests telemetry data from the monitoring services your clinic already uses, normalizes it across vendors, scores it by clinical urgency using device-type-aware logic, and presents a single prioritized queue to your clinical team each morning.

It doesn't replace Medtronic CareLink, LATITUDE, Merlin.net, or Biotronik Home Monitoring. Those platforms remain the authoritative source of device data and the manufacturer-to-patient communication channel. Implansense reads from them; it doesn't replace them.

It doesn't replace the EHR. Clinical documentation, order management, and the patient's longitudinal medical record live where they've always lived. The Clinical Timeline in Implansense provides EHR-adjacent context — device events, transmissions, follow-up interactions — in a format optimized for device clinic workflow, not for replacing the medical record.

And it doesn't make clinical decisions. The triage scoring surfaces urgency; the clinical judgment about what to do belongs to the clinician. That distinction matters and it shapes every feature decision we make.

Philadelphia and what comes next

We're building Implansense in Philadelphia because Philadelphia is a genuine medtech ecosystem — Penn Medicine, Jefferson Health, Temple Health, and a network of community cardiology programs representing a significant and diverse device clinic population. The opportunity to build in proximity to the clinical teams we're serving has shaped the product in ways that remote development wouldn't have.

In early 2026, we closed a $3.5 million seed round that will let us accelerate platform development and deepen integrations with the remote monitoring services and EHR systems that our clinic partners use. The funding reflects what we've heard consistently from cardiology IT directors, EP program directors, and device clinic nurse managers: the problem we're solving is real, it's costing clinical programs meaningful time and carrying meaningful patient safety implications, and no one else is focused specifically on the cross-vendor clinical triage layer.

We started with the observation that a nurse manager couldn't tell at 9 AM which of her morning's transmissions was the one that needed her first. Everything we've built is designed to change that.