# How Chips Use On-Chip Pressure Sensors for Medical Implants

An episode of [The Hardware Podcast with Fexingo: Chips, Devices, and Electronics Engineering Conversations](https://invisiblenetwork.ai/podcast/1896799982), 10 Jul 2026, 8:31.

Audio: https://audio.fexingo.com/business/the-hardware-podcast/episode-0102.mp3

This episode dives into the growing role of on-chip pressure sensors in implantable medical devices. Lucas and Luna discuss how microelectromechanical systems (MEMS) pressure sensors are being integrated directly into silicon for applications like continuous intraocular pressure monitoring in glaucoma patients and smart cardiovascular stents. They examine the key engineering trade-offs: miniaturization vs. sensitivity, long-term drift compensation, and the challenge of powering these sensors without batteries. A recent study from the University of Michigan demonstrated a chip that can measure intracranial pressure with 0.1 mmHg accuracy for over a year. The conversation also touches on how these sensors communicate wirelessly with external readers, and why the shift from discrete components to fully integrated sensor chips could reshape the $500 million implantable sensor market by 2028. #OnChipPressureSensor #MEMS #ImplantableMedicalDevice #Glaucoma #IntraocularPressure #CardiovascularStent #WirelessSensor #BiomedicalChip #Semiconductor #MedTech #UniversityOfMichigan #IntracranialPressure #SensorIntegration #Technology #Engineering #HardwarePodcast #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo

## Transcript

Lucas: So last month at the IEEE International Solid-State Circuits Conference, there was a paper from a team at the University of Michigan that really caught my attention. They demonstrated a single-chip pressure sensor that can monitor intracranial pressure continuously for over a year, with accuracy down to 0.1 millimeters of mercury.
Luna: That's impressive. But what makes that different from the pressure sensors we've had for decades in things like tire pressure monitoring systems?
Lucas: Right, great question. The key difference is integration and power. Traditionally, a pressure sensor in a medical implant would be a discrete MEMS element wired to a separate ASIC for signal processing, and then you'd need a battery pack. This Michigan chip puts the MEMS diaphragm, the analog front-end, the ADC, and a wireless transmitter all on the same piece of silicon.
Luna: And the battery?
Lucas: No battery. It uses inductive power harvesting from an external reader, so the patient wears a small patch or holds a wand near the implant. That eliminates the need for surgical battery replacements, which is a huge deal for chronic conditions.
Luna: Okay, so let's talk about the actual use cases. Glaucoma comes to mind.
Lucas: Exactly. Glaucoma is the second leading cause of blindness globally, and the primary risk factor is elevated intraocular pressure. Right now, patients get their pressure measured during clinic visits, maybe every few months. That's a snapshot. Pressure can spike at night or after certain activities, and you miss it.
Luna: So an implant that reads continuously could give doctors a much clearer picture.
Lucas: Precisely. A company called Implandata Ophthalmic Products actually has a ce marked implant called the Eyemate that does this. It's a ring-like device placed in the eye during cataract surgery. But that's a multi-chip module — their current version is about 7 millimeters in diameter and 1 millimeter thick.
Luna: And the Michigan chip is smaller?
Lucas: Significantly. Their prototype is under 3 millimeters on a side, and they believe they can shrink it further. That matters because you want to minimize the surgical footprint and the foreign-body response.
Luna: What about drift? I've read that MEMS pressure sensors tend to drift over time as the diaphragm material relaxes or biofouling occurs.
Lucas: That's the engineering challenge. The Michigan team addressed it by integrating a reference cavity that is sealed at a known pressure, and they periodically compare the sensor reading to that reference. They also use a compensation algorithm that tracks temperature and humidity inside the package. Their paper reports less than 1 millimeter of mercury drift over 14 months of continuous operation.
Luna: That's remarkable. Now, I know the cardiovascular space is also interested. Stents that can sense pressure changes to detect restenosis?
Lucas: Exactly. There's a startup called Vectorious Medical Technologies that has a wireless microsensor mounted on a stent. It measures left atrial pressure in heart failure patients. When pressure rises, it's a sign of fluid overload, and the patient can adjust their diuretics early, preventing hospitalization.
Luna: That's the kind of closed-loop medicine that could really cut costs. But these are still discrete sensors, right? Not fully integrated on-chip.
Lucas: Right now they are. But the Michigan work shows the trajectory. I think within five years we'll see commercial implants with the sensor and the entire readout chain on a single chip.
Luna: You know, it's conversations like this that remind me why we keep this podcast ad-free. We don't have to pander to a sponsor who wants us to hype up a product that isn't ready yet.
Lucas: That's exactly the idea. We can talk about the real engineering, the trade-offs, the stuff that matters. And if listeners find value in that, they can support the show at buy me a coffee dot com slash fexingo.
Luna: Yeah. No ads, no fluff. Just the details. So back to the chip — what about the wireless link? How do they get the data out?
Lucas: Most of these designs use near-field communication at 13.56 megahertz, the same frequency as RFID. The external reader provides power and the chip backscatters data by modulating its impedance. Data rates are low, a few kilobits per second, but you don't need high bandwidth for a pressure reading every few minutes.
Luna: What about the antenna? If the chip is that small, the antenna must be tiny.
Lucas: That's another integration challenge. Some designs use a planar coil that wraps around the chip perimeter. Others use the silicon substrate itself as a ground plane and a deposited metal layer as the antenna. The Michigan chip uses an on-chip coil that's about 2 millimeters in diameter, which limits the read range to about 5 centimeters. That's fine for a bedside monitor.
Luna: I imagine there are also concerns about biocompatibility and hermetic sealing.
Lucas: Absolutely. The chip has to be coated with parylene or silicone to prevent bodily fluids from shorting out the electronics. And the MEMS diaphragm has to remain exposed to the fluid to measure pressure. So you need a selective coating process that protects the circuitry but leaves the sensing element open. That's nontrivial.
Luna: Let's talk about market size. Who's investing in this?
Lucas: The global market for implantable pressure sensors was about 500 million dollars in 2025, growing at a compound annual growth rate of around 12 percent. The big players include Medtronic, Abbott, and Boston Scientific. But there's also a wave of startups like the ones we mentioned, plus academic spinouts.
Luna: And what about the regulatory path? The FDA has a separate classification for implantable sensors.
Lucas: Right, most of these devices are Class III, so you need a premarket approval application. That means extensive clinical trials. The Michigan chip has only been tested in animals and cadaver models so far. But the underlying MEMS technology is mature — it's been used in automotive and industrial applications for decades. The novelty is the integration and the wireless power.
Luna: So the risk is more about surgical placement and long-term reliability than the chip itself.
Lucas: I'd say so. And also the business model. You need to convince surgeons to adopt a new implant, and hospitals to pay a premium for a sensor-enabled device that might reduce readmissions. The cost benefit is clear for conditions like heart failure, but adoption takes time.
Luna: One more question: could these chips be used for drug delivery? Like a closed-loop insulin pump with pressure sensing?
Lucas: There's research on that. You can use a pressure sensor to detect when a drug reservoir is empty or if there's a blockage in the microfluidic channel. And you could integrate a micropump on the same chip. That's a harder integration problem, but some groups have demonstrated it.
Luna: So the future is truly multi-functional chips that sense, process, communicate, and actuate.
Lucas: Exactly. And I think the pressure sensor is just the first wave. Once you have a reliable way to power and communicate with a chip inside the body, you can add other sensors — temperature, glucose, even neural recording. The Michigan lab is already working on a multi-sensor platform.
Luna: That's exciting. But let's be realistic: we're still years away from widespread clinical adoption.
Lucas: True. But the rate of progress in on-chip integration is accelerating. Ten years ago, a chip like this would have been science fiction. Now it's a prototype. I suspect by 2030, we'll see the first fully integrated implantable sensor chip cleared by the FDA.
Luna: And that will change how we manage chronic disease.
Lucas: Absolutely. It shifts medicine from episodic to continuous. And that's a fundamental shift.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.