Medical care needs to become much more personalised. A standard treatment doesn’t work for everyone. But how do you determine which treatment is best for an individual patient, especially when dealing with large and complex datasets? That’s the key question driving Saddle Point Science Europe, a spin-off from King’s College London, now based in Nijmegen.
The company develops advanced data analysis tools that make medical predictions smarter, more accurate, and more tailored. “We constantly ask ourselves: what analytical methods are truly needed in medical care?” says Ton Coolen, founder and Professor of Neurophysics at Radboud University.
When Ton’s wife became ill, he started volunteering at London’s largest cancer hospital. “I discovered that my background in mathematics was more useful for analysing medical data than for reviewing budgets.” He worked on studies, including breast cancer research, analysing how tumours behave across different patient groups. “Sometimes tumours may appear similar but respond completely differently to treatment. Thanks to new statistical analyses, we were able to uncover those differences, helping doctors decide who will genuinely benefit from a certain treatment – and who won’t. It helps avoid unnecessary side effects and increases the chance of success.”
Ton founded his first company in London, Saddle Point Science Ltd, to gain more freedom in his work. After Brexit, he moved to Nijmegen and started Saddle Point Science Europe. “We still collaborate closely with London, but now the main focus is here.”
Old methods, new challenges
Saddle Point Science develops new mathematical and statistical methods for analysing medical data. These are necessary because, as Ton explains, the medical field often still uses analysis techniques dating back to the 1970s. “These methods were based on the assumption that one has a substantially higher number of patients than data points per patient. In the past, you would indeed typically measure only a handful of features per person – such as blood pressure, age, and tumour size. Many traditional statistical models are based on that setup.”
Today, however, the situation is reversed. Researchers can now measure thousands of features in a single patient – from DNA profiles to protein levels and molecular structures. This creates datasets with far more variables than patients. Classical methods struggle in that context – they simply weren’t built for such complex, high-dimensional data.
That’s why Saddle Point Science develops software tools like spsSIGNATURE, which helps make reliable predictions based on complex datasets, and spsMOSAICS, which identifies hidden patient subgroups. “We aim to support doctors in answering questions such as: who will likely respond well to this treatment? Should someone receive aggressive treatment?”
Bridging science and practice
The company works closely with physicians and medical researchers. “We ask them what challenges they face in practice, then use our knowledge of mathematics and physics to design solutions.” According to Ton, the team’s strength lies in that bridge-building role. “We speak the languages of both theoretical physicists and medical professionals. That combination is rare – but essential.”
And it shows in the application. Saddle Point Science’s software is already being used in European research projects and by pharmaceutical companies such as GSK in London. “We license the software only to partners who understand how to use it. It’s a powerful technology, but it’s not a black box.”
Growth and collaboration
The Nijmegen-based team is growing. There are now four people, soon to be five. Ton combines his work at the company with his professorship at the university, where he supervises students and connects Master’s projects to real-world challenges. “It’s a great synergy. We’ve already been able to recruit two people straight from their studies.”
To accelerate growth, Ton turned to Briskr and the Business Angels Network Nijmegen. “We joined a pitching programme and received coaching from an experienced TEDx speaker. It was intense but incredibly valuable. Afterwards, we got the chance to pitch to investors, which led to useful connections and gave us detailed and constructive feedback.”
Global ambitions, personal impact
The company has big ambitions. “From Nijmegen and London, we want to expand across Europe and then into Japan and the United States, where we already have some contacts. If, in five years, just 1% of our potential users are working with our software, our impact will be huge.”
And what makes the work so meaningful? Ton doesn’t hesitate: “Improving medical care by using data more intelligently – that’s what it’s all about. We want to help ensure that new treatments reach the right people faster. That benefits science, the healthcare system – and most of all, the patient.”
