Volpara Health: The Algorithm, the Breast, and the Belief

Volpara Health exists to save families from cancer.

Founded in Wellington and now used in clinics worldwide, Volpara provides physics-based, FDA-cleared breast-health software that helps clinicians assess cancer risk, optimise image quality, and support more personalised screening decisions. Over more than a decade, the company scaled from early grants to public listing, U.S. expansion, and two acquisitions.

A Photograph, a Lightbox, and a Realisation

In the 1990s basement of Oxford’s Robotics Research Group, Ralph Highnam and colleagues taped mammography films to lightboxes and photographed them with robot cameras. On a screen, the same breast image looked starkly different depending on the lighting.

“We realised if you just throw machine learning at raw images, it won’t be reliable,” Highnam recalls.

The fix would be heretical for the AI zeitgeist for decades: standardise the physics of the image first, then learn from it. That decision—normalise before you optimise—became the technical spine of Volpara Health.

By 2009, digital mammography was finally mainstream. Highnam—by then in Wellington—re-formed an old alliance with Sir Mike Brady (Oxford), Nico Karssemeijer (Nijmegen), and Martin Yaffe (Toronto). The idea was simple and audacious: turn decades of breast-density research into robust, clinic-ready software that would help personalise screening for millions of women.

Volpara’s first spark was funded not by VCs but by micro-grants: a NZ$10k seed from Grow Wellington, support from Callaghan Innovation, and donated IP legal work.

That scaffolding let Highnam translate physics into product.

Ralph x Angus: The Volpara Health Top 10

Top 10 takeaways from Volpara founder Ralph Highnam, in conversation with Angus Blair, General Partner at Outset Ventures.

Context — Why density mattered, and why timing did too

Breast density is clinically tricky: dense tissue and tumours both appear bright on a mammogram, making cancers harder to see. The concept had circulated for years; what didn’t exist was a standardised, automated way to measure it at scale and make it legible to clinicians in the moment.

Volpara’s bet was that physics-grounded algorithms—quantifying breast composition—would travel better across vendors, hardware quirks, and environmental artefacts than end-to-end black-box models. The market inflection—digital imaging everywhere, clinics hungry for workflow-ready tools, and a growing regulatory consciousness—made 2009 a now-or-never window.

The Spark — Clean cap tables, clean data, clear intent

Volpara wasn’t Highnam’s first attempt to commercialise imaging science. A decade earlier, his Oxford spin-outs OMIA and OXIVA had merged to form Mirada Solutions—one of the university’s earliest AI-in-medicine startups, eventually sold to CTI and later absorbed by Siemens. That journey left him with a durable operating philosophy for how to commercialise deep tech without losing control.

Rule one: Keep the company clean.

Rule two: Build a quality system early.

Rule three: Get independent validation before you fall in love with your own graphs.

Rule four: Sequence your capital with the same discipline.

Early angel money came from the same Oxford network that had backed Mirada. Around 2012, John Hood—former Oxford Vice-Chancellor—and Roger Allen led a NZ$2 million round motivated as much by mission as by maths. In 2014, Neil Craig and Craigs Investment Partners raised a NZ$5 million pre-IPO bridge that prepared Volpara for its public debut.

Each round had a distinct purpose: grants to build, angels to validate, institutions to scale.

WellyForge

Wellington's Collision of Science, Tech & Innovation

Ralph brings together science & city to create a more connected Wellington via a weekly newsletter celebrating local successes and a monthly meet up.

Subscribe now!

The Turning Point — Make the science legible

Early on, Volpara’s output was a beautifully scientific scorecard—and borderline unusable in a busy clinic. A US business-development lead returned from a physician meeting with a blunt request: translate it to A, B, C, D. Four weeks later, under trade-show pressure, the team shipped a mapping clinicians could recognise at a glance. It wasn’t final, but it was good enough to unlock pilots and sales.

At first installs, things broke. One US hospital’s PACS nearly melted under a flood of duplicate messages thanks to an innocent retry loop. The fix wasn’t just code—it was responsiveness. Soon came a more positive surprise: a California site found they could have density conversations while the patient was still in the clinic, catching more cancers and billing appropriately. They offered to pay—Volpara’s first US$50k-per-year willingness-to-pay signal.

“We learned to ship for workflow, not just science. The right abstraction turns evidence into action.”

Distribution — Logos for credibility, rights for survival

Highnam is adamant: never go exclusive on distribution. Volpara signed GE early—not to outsource sales, but to place a marquee logo in investor decks.

“You can sign the logo; just don’t sign away your right to learn from customers.”

The team insisted on direct-sale rights because missionary sales—educating a sceptical market—can’t be subcontracted to a field rep juggling three other products.

Two unglamorous lessons:

1. Category creation takes years before repeatability.
2. US first, not Europe first. CE approval was administratively easier but buyers slower; American clinics moved faster on new tech once workflow value was proven.

Regulatory & Reimbursement — “Good enough” FDA, and value beyond codes

Volpara didn’t wait for a perfect algorithm to chase FDA clearance. The company filed when the science was defensible and improvable post-clearance. That sequencing both de-risked the business and converted investor objections into a data plan, not a theology debate.

On reimbursement, Volpara went contrarian: rather than chase an existing billing code, it sold operational value already present in the workflow—follow-on imaging, objective documentation that reduced audit pain, and quality metrics.

First FDA clearance came in 2010–11; first U.S. revenues followed within a year, proving the model before scale.

Assessing breast density is hard: AI can help

The Capital Journey — From Angels to the ASX

Volpara’s path to global scale wasn’t just technical—it was a financial education in sequencing risk.

• 2009–2010: Seeds of Support — Micro-grants and local mentors kept the lights on while the algorithm matured.
• 2012: The Mission-Driven Round — NZ$2 million from John Hood and Roger Allen turned goodwill into growth capital.
• 2014: Preparing for the Leap — A NZ$5 million raise with Craigs Investment Partners funded U.S. expansion and SaaS transition.
• 2016: Listing on the ASX — The IPO raised A$10 million, with another A$10 million soon after. Highnam, once sceptical of “hyper-smooth public CEOs,” found the discipline of quarterly reporting sharpened focus.
• Post-IPO: The SaaS transition hurt in the short term—retraining a capital-sales org and educating investors—but recurring revenue forced deeper listening and ultimately stability.

“SaaS hurt at the start and saved us later. Recurring revenue forces you to listen.”

• 2024: Acquired by Lunit Technologies — Lunit Completes Acquisition of Volpara: Pioneering Next-Level of AI-Driven Cancer Care

Engineering Culture — Sustainable speed, physics + AI, and saying “no”

Volpara’s technical stance never chased fashion. In the deep-learning boom, it kept a physics-first, AI-second architecture: compute the volumetric truth, then apply learning. When competitors trained on radiologist labels to predict A–D directly, Volpara’s team worried about “ground truth”—human subjectivity.

“If your labels are subjective, your model inherits that subjectivity and won’t travel.”

Culture mattered as much as code. Volpara built with a New Zealand cadence—high-quality, family-friendly hours—paired with U.S. commercial muscle. People moved to Wellington for a mission: “Saving families from cancer.”

The company designed for robustness: no site-by-site calibration, minimal configuration, instant-read outputs.

When a vendor’s new tilting paddle broke assumptions mid-deployment, the team patched fast and re-architected—proof that robustness beats heroics.

“If you have to calibrate every site, it won’t scale. Design for robustness, not heroics.”

Resolution — Principles that travel

By the time Volpara listed, its density software was in clinics across continents. The company broadened into a SaaS breast-health platform, leveraged real-world data for quality metrics, and kept investors close with transparent clinical and regulatory reporting. Highnam later handed the CEO role to a scale-optimised leader—an intentional, not reactive, handover.

Closing Reflection — The courage to be unfashionable

Volpara’s origin story isn’t one of flawless foresight—it’s a chain of principled choices made under constraint: clean up the IP, build the QMS early, translate science for humans, take the FDA shot when it’s good enough, refuse exclusivity, keep physics in the loop, and fund step-by-step without surrendering control.

In a decade when “just throw deep learning at it” was the most seductive sentence in tech, Volpara bet on standardisation, scepticism, and patient-centred value.

Progress wasn’t a straight line from novelty to scale—it was a discipline of gates.

The company that reached millions wasn’t the one that chased fashion; it was the one that made itself testable, legible, trustworthy—and capital-ready—again and again until the market couldn’t ignore the results.

The Founder

Ralph Highnam is the founder and former CEO of Volpara Health, the New Zealand med-tech company that transformed breast-density physics into software now used in clinics worldwide. Over 13 years, Ralph led Volpara from a Wellington start-up through FDA clearance, ASX listing, and two U.S. acquisitions. He now splits his time between community projects like WellyForge and WellyBreastHub, mentoring deep-tech founders, and serving as Adjunct Professor at Victoria University of Wellington.

The Interviewer

Angus Blair is a General Partner at Outset Ventures, New Zealand’s leading deep-tech incubator with an active seed fund. He’s backed 25+ science-led spinouts in aerospace, advanced materials, med-tech, and fusion — writing first cheques from $100k to $2m and following through scale-up.