Cleared, But Not Adopted: SpectraWAVE's Former CEO on Why Proof Looks Different at Every Stage

Contents

1. What the episode covers
2. The main points
3. Going further
4. ”The company got FDA clearance — why is revenue still flat?”
5. ”They got acquired — good outcome, or forced sale?”
6. ”How do I check what management says?”
7. Worth a look
8. One thing to take with you

To those above the average, you may speak of higher things; to those below the average, you may not speak of higher things.

—— Analects, Book VI (Yong Ye), Confucius (Spring and Autumn period; translation mine)

In the 21 September 2026 episode of Medsider, host Scott Nelson interviewed Eman Namati, former CEO of SpectraWAVE. The company builds imaging systems used inside coronary arteries; Philips acquired it this year and Namati now leads advanced imaging there. The line that stayed with me was about evidence: what gets you a regulatory clearance and what gets a physician to change how they work are two different things — he spent a decade at an earlier company accumulating nine clearances without one that carried a strong diagnostic claim. That experience sits on the US 510(k) pathway; drugs and other markets set their own bar.

What the episode covers

Medsider is a show Scott Nelson has been running for over a decade, interviewing founders and CEOs of medical device and health technology startups. Namati holds a PhD in medical imaging and biomedical engineering, did postdoctoral work at Harvard Medical School, spent time at the University of Iowa and Massachusetts General Hospital, then moved into industry — rising from principal scientist to CEO at Nine Point Medical before building SpectraWAVE.

A quick note on the products, since the later discussion leans on them. The flagship is HyperView, which stacks two technologies on one catheter: optical coherence tomography (OCT), which uses light to image the structure of the artery wall in fine detail, and near-infrared spectroscopy (NIRS), which reads chemical composition to tell you what the plaque is made of. Together they let a physician decide whether to place a stent, where to place it, and whether the job is done. The second line, X1 FFR, estimates how much a narrowing restricts blood flow from a standard X-ray angiogram, skipping the pressure wire.

Inside Philips, both landed in a team that already leads two categories — digital intravascular ultrasound on the imaging side, the OmniWire pressure wire on the physiology side. Namati describes the imaging pair as covering opposite ends: ultrasound is simple and fast for daily work, HyperView delivers the higher resolution and fuller plaque assessment.

The main points

1. Moving from scientist to CEO means giving up “the best technology” as the goal. The best technical solution and the best product often diverge, he says: the technical winner can be expensive, hard to manufacture, hard to use, disruptive to the workflow, or slow to build. Running a company means optimizing for far more than the technology. The other thing to give up is the single-author game of academia — first authorship, principal investigator, everything orbits one person. Building a machine a hospital will buy takes a team of strong people across many disciplines.

2. Deciding how much more information to gather is itself one of the critical decisions. A scientist’s instinct is to collect more data and answer every question fully. As CEO, he found himself making consequential and sometimes unrecoverable calls on limited information — which makes “is this decision worth another three months of data” a skill in its own right.

3. Every stage gets its own minimum viable product. People in medtech often say MVP doesn’t apply, since the thing goes into patients. His framing is per-stage: the device you take into first-in-human should look different from the one in a pivotal trial, which should look different from the one you scale. Early versions need differentiation without deep integration into the broader ecosystem, designed so you can come back and build on them. Manufacturability, serviceability and cost arrive at scale-up. Two hard limits: safety is never compromised at any stage, and don’t walk through a door you can’t walk back through.

4. Preclinical work surfaces what the bench missed, every single time. This is the thing that kept surprising him. However thoroughly a team vets something on the bench, every preclinical study turns up something new on the engineering side — usually not physiology, just a gap nobody saw. His phrase: a preclinical study sharpens the sword differently than a bench study does. Scott added his own version: a well-known venture firm once pushed his team to run a feasibility study the FDA wasn’t requiring. His instinct was to minimize clinical work; they ran it anyway and were glad they did.

5. Clearance evidence and adoption evidence differ by an order of magnitude. The 510(k) path turns on substantial equivalence to a predicate plus safety and efficacy — which sounds like it should be enough. His observation is that most startups on that path end up with relatively light claims, and in the diagnostic setting especially so. Changing physician behavior takes multi-center, multi-year studies, plus reimbursement, plus guideline change. Nine clearances over ten years at his prior company spanned initial release, new features and new indications, and still none of them carried a strong claim about diagnostic capability. So the question to answer first: is this evidence to get into the market and find the right customer, or evidence to drive adoption? He also flags that reimbursement expertise rarely lives in-house at a startup, outside firms do this well, and the timelines for reimbursement, evidence and clearance have to be synchronized — none of them are short.

6. Over half of investor diligence is about whether they trust the CEO. From first introduction to closing a round, six months is excellent, nine is typical, twelve happens. Investors check the numbers and the technology, but his estimate is that more than half the weight sits on trust — you said you’d do A, you came back and said A is done, now we’re doing B. That record is built one interaction at a time, which is why conversations with investors who won’t join this round still pay: they may lead the next one, and watching from the sidelines completes half their diligence. His first chairman told him early on: we’re playing to win. Be efficient, but if spending another ten or twenty percent of the raise is the difference between success and failure, nobody complains about that 0.2x once the return is three, five or ten.

7. Build for independence, not to be bought. This is the lesson he says he repeats most. You don’t control an acquisition; too many outside factors drive it. Inside SpectraWAVE, from him down, everyone was building toward an independent company at scale. That north star changes daily decisions — outsource this or build it in-house, cut this corner to save two months or don’t. Owning it end to end costs time and effort, and buys you two things: you learn far more about the technology, the product and the space, and when trouble comes you have more optionality. When a liquidity opportunity does appear, put your best foot forward — just don’t hang the company’s direction on it.

Going further

”The company got FDA clearance — why is revenue still flat?”

I’ve been caught by this one. Clearance hits the wire, the stock pops, and two years later revenue has barely moved while shareholders start questioning management’s execution. After this episode I’d split the question in two.

First: what is the clearance itself worth? The 510(k) logic is equivalence to an existing device, so the claims you walk away with can be narrow — narrow enough that a physician reads them and thinks their current tool already does that. A clearance is a ticket to the venue. The price of the ticket tells you nothing about the revenue inside.

Second: are the things adoption requires actually in motion? He named three — large clinical evidence, reimbursement, guideline change. Each takes years, and they have to line up: without evidence the payers won’t move, without payment the hospital can’t make the math work, and without a guideline most physicians won’t rebuild their workflow around a new machine. So when I read a clearance announcement now, I go looking for where those three stand: an enrolling multi-center study, a pending reimbursement code, a society position statement that mentions the category. If none of that exists, the clearance deserves a heavy discount in any valuation.

It also explains a familiar frustration: the count of cleared products on a company’s slide keeps climbing while the revenue line stays flat. Nine clearances that don’t add up to one strong claim was lived experience for him; on a deck it renders as a tidy row of milestones.

”They got acquired — good outcome, or forced sale?”

I used to read acquisition news by looking at the price and the premium. His independence argument gave me another angle: look at which logic the company was built on before the buyer showed up.

A company built to be bought takes shortcuts in specific places — manufacturing outsourced, no direct channel, a team sized to tell a story. Those choices look smart on the day of the deal, because the acquirer brings all of it. If no buyer appears, the same company is stranded somewhere it can’t move forward from, and there’s exactly one option left on the table.

A company built for independence holds manufacturing, channel and clinical relationships itself, so an acquisition becomes upside. That’s how I read his point about optionality: you only get to choose when you can survive alone. The same test works on a holding — when consolidation news hits a sector, ask what this company looks like in three years if nobody buys it. If you can answer, the news is an option. If you can’t, it’s a countdown.

”How do I check what management says?”

Listening to him describe how investors build trust in a CEO, I kept thinking the method transfers straight to individual investors, at no cost.

It works like this: write down the commitments from one earnings call — what ships in which quarter, how many new sites, where gross margin is heading — and file it. Next call, reconcile the old list before listening to the new story. That’s what a venture investor spends twelve to eighteen months doing from the sidelines. They need to book meetings; I need a text file.

The biggest payoff since I started doing it is that I can now tell two kinds of company apart. One closes the loop every time — done is reported as done, missed is explained. The other keeps introducing new things while old promises quietly vanish. The second kind can still post decent numbers, and I sleep worse owning it, because I have no way to check it against its own words. The habit also makes a single piece of good news easier to size: dropped into a two-year sequence of promises, it often shrinks into something they’d already announced three times.

Worth a look

• The Medsider episode itself, 21 September 2026, Scott Nelson with Eman Namati. The full interview archive and the free courses now live at scottnelsonlive.com.
• SpectraWAVE’s product pages at spectrawave.com, covering HyperView and X1 FFR, now under Philips.
• Two people mentioned in passing: Bob Paulson, who has built startups that had to secure new CPT codes and whose episode Scott recommends for anyone digging into reimbursement, and Bill Hoffman of Inari, cited as the standard-bearer for building a company you intend to commercialize yourself.
• For background, the difference between the 510(k) and PMA pathways is worth twenty minutes — much of this conversation rests on that fork.

One thing to take with you

The proof that gets you through this gate and the proof that changes someone’s behavior are two separate things, and you have to prepare them separately.

What landed hardest in that episode was this: nine clearances didn’t buy one behavior change, and the reason wasn’t insufficient effort. Those nine efforts were answering a different question.

Here’s something I’ve tried that takes under ten minutes. Pick something you’re currently trying to convince someone of — getting your parents in for a cardiac screening, getting a manager to adopt a new process, convincing yourself to keep holding a stock. Draw two columns. Left: what it takes to clear this gate. Right: what it takes for them to actually change what they do. The left column is usually one line — the doctor says fine, the manager nods, the earnings didn’t break. The right column has to be concrete: your parents need a real example from someone their age, your manager needs the hours the next team saved, you need two consecutive quarters of revenue tracking what management said.

You’ll likely find the two columns don’t match, and that most of your effort has gone into the left one. Over the next month, pick the cheapest item in the right column and do that one.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.