News
Eurostars winnaarsbijeenkomst
NEUROrehabilitation with TWIN-based Prediction and Optimization at Scale
We were honoured to be invited to the Eurostars Winners’ Meeting for our NEUROTWINS project – a hyper-scalable, data-driven stroke rehabilitation solution that bridges hospital and home care. By transforming everyday devices such as smartphones, laptops, and TVs into powerful telerehabilitation tools, NEUROTWINS aims to improve patients’ quality of life, personalize recovery programs, and significantly reduce healthcare costs.
Together, these projects reflect our mission to deliver cutting-edge, impactful innovations that address some of the world’s most pressing challenges.
MIT haalbaarheidsproject
Our team has received a grant to conduct a feasibility study into innovative quantum computing technology. Quantum computing harnesses the unique property of atomic particles to exist in multiple states simultaneously. This enables extremely complex calculations – which would take impractically long on conventional computers – to be performed much faster.
Among the two most common approaches to quantum computing – quantum annealing (QA) and gate-based quantum computing – QA was the first to be commercialized. Interestingly, the mathematics of QA shows remarkable similarities to quantum machine learning, which is considered the next generation of quantum computers.
With this study, we aim to explore how new algorithms can extend the scope and effectiveness of calculations on existing quantum computing hardware. Such algorithms are an important step toward the functional application of quantum computers in areas such as drug discovery, the development of new materials, and complex financial modeling.
ENMC Workshop
On March 27- 29, 2025, the 287th European Neuromuscular Centre (ENMC) workshop took place in Hoofddorp, the Netherlands, focusing on the harmonization and federated analysis of Myotonic Dystrophy (DM) registries to model heterogeneous disease trajectories. The workshop gathered clinicians, researchers, and patient representatives to discuss the future applications of harmonized and federated natural history data, emphasizing methods to model personalized disease trajectories under standard care and their use as individualized baselines for clinical trials.
A presentation was given by Ton Coolen on Saturday where he discussed "New Statistical Tools for Personalized Medicine" He highlighted advanced statistical techniques, including Bayesian approaches, to uncover underlying heterogeneities in disease progression, supporting more accurate predictions and personalized treatment strategies.
Health Valley
Exciting news! On March 13rd, we had the incredible opportunity to attend the Health Valley Event in Nijmegen, and we are truly grateful for the amazing connections we made. It was a fantastic experience showcasing our innovative solutions and engaging with so many inspiring individuals. A big thank you to everyone who stopped by our booth—we appreciate your time and enthusiasm! Looking forward to staying connected!
BANN
On March 10th, we were invited to pitch our company to the Business Angel Network (BANN) in Nijmegen! It was an inspiring experience to connect with angel investors, share our vision, and discuss our innovative solutions. We are grateful for the valuable insights, support, and connections we gained during the event. We are grateful for this opportunity and are eager to see what the future holds!
EORTC
In February 2025 we started a strategic collaboration with EORTC in Brussels, aimed at testing the ability of our analytics pipelines to identify the responders in failed clinical trials. The phase 3 trial we will initially focus on aimed at finding out whether adding marizomib, a proteasome inhibitor that crosses the blood–brain barrier, to standard therapy prolongs the survival of patients with newly diagnosed glioblastoma. The answer was negative (see Roth et al, 2024, Neuro-Oncology 26(9), 1670–1682). Now it will be exciting to see, in collaboration with our partners at EORTC, whether with more advanced analysis of these trial data a responder sub-group of patients can be found!
GSK
We are delighted to announce that from January this year we are collaborating closely with the Oncology Translational Research Division of GSK in the United Kingdom.
GSK will be using our data analytics pipelines spsMOSAICS and sposSIGNATURE to increase the efficiency of their cancer drug discovery pipelines and trials. This should lead to more precise and reproducible drug response predictions and faster identification of promising therapeutic componds, and thereby accelerate the process of getting new personalized cancer treatments to the patients that need them!
Eurostars grant
SPARCS (Science-based, Personalized Aphasia Rehabilitation Continuum after Stroke) is a scalable VR-based telerehabilitation solution for aphasia (impairment in a person's ability to comprehend or formulate language because of damage to specific brain region) after stroke, enabling remote, multi-patient therapy while reducing costs and improving care. Using XAI-driven patient models and head-mounted displays, it predicts treatment responsiveness and optimizes recovery protocols. With stroke affecting 81M people globally and healthcare systems under strain, SPARCS addresses the urgent need for accessible, high-quality telemedicine solutions to prevent post-discharge deterioration and enhance patient outcomes. We are excited to work together with Eodyne Systems (Spain) and with Centre Hospitalier Universitaire Vaudois (CHUV) (Switzerland).
MIT grant
Sapiens and SaddlePoint are collaborating on BReIn-VR to develop a multiplayer VR game for stroke rehabilitation. This interactive platform focuses on speech, cognitive functions, and hand-arm movements, integrating AI-driven data analysis and predictive modeling. The project aims to enhance patient recovery and independence through two key objectives:
-
Developing specialized rehabilitation training in a multiplayer VR environment.
-
Creating transparent AI algorithms to monitor and predict patient recovery.
These AI models will provide both doctors and patients with better insights, enabling real-time adjustments for personalized care. Expected outcomes include scalable AI-driven predictions, explainable analyses, real-time data collection, and specialized VR rehabilitation games.
Figon
Ton Coolen was attending the FIGON Dutch Medicines Days on behalf of Saddle Point Science Europe the 16 and 17th of October, showcasing our poster entitled "𝘞𝘩𝘰 𝘢𝘤𝘵𝘶𝘢𝘭𝘭𝘺 𝘣𝘦𝘯𝘦𝘧𝘪𝘵𝘴 𝘧𝘳𝘰𝘮 𝘢 𝘥𝘳𝘶𝘨? 𝘛𝘰𝘰𝘭𝘴 𝘧𝘰𝘳 𝘥𝘢𝘵𝘢-𝘥𝘳𝘪𝘷𝘦𝘯 𝘱𝘳𝘰𝘴𝘱𝘦𝘤𝘵𝘪𝘷𝘦 𝘳𝘦𝘴𝘱𝘰𝘯𝘥𝘦𝘳 𝘪𝘥𝘦𝘯𝘵𝘪𝘧𝘪𝘤𝘢𝘵𝘪𝘰𝘯." Where he was explaining about the innovative approaches to data analytics Saddle Point Science apply in precision medicine.
Seminar
Ton Coolen was invited to the European Organisation for Research and Treatment of Cancer (EORTC) in Brussels on June 25th, to give a seminar on `Prospective responder identification, overfitting correction and Bayesian Federated Inference’. EORTC is a prestigious independent research organisation with significant expertise in the initiation, design and execution of cancer trials. Ton’s visit was used also to discuss opportunities for collaboration between EORTC, Saddle Point Science Europe, and Radboud University. One of the key topics of mutual interest identified is the use of novel analytical techniques for the prospective identification of who will be the responders to certain therapies. We are very much looking forward to working with EORTC on this!
PersOn
Explainable, Maintainable, and Trustworthy Decision Support Systems for Personalised Care in Oncology
The research carried out within this NWO-funded consortium proposes to improve cancer patient care by developing advanced clinical decision support systems (DSS). Current systems often fail to meet patient expectations for personalized treatment and lack effective explanations, flexibility, and integration into clinical workflows. The goal is to create next-generation DSS that enable shared decision-making in oncology, thereby enhancing patient-centered care and optimizing the use of AI tools in clinical practice.
Joint HMA/EMA workshop
On Friday June 14th 2024, Saddle Point Science Europe participated in EMA’s Joint HMA/EMA workshop on real-world evidence (RWE) methods in Amsterdam. Real-world data are typically collected from hospital or GP registries, and relate to drugs or treatments that have already been licensed following standard clinical trials. They play a significant role in fully understanding the effectiveness and safety of medications.
Topics discussed included regulatory decision-making, ranging from non-interventional studies to utilizing real-world data as external controls in clinical trials to support marketing authorizations. The primary takeaway of the meeting was emphasis of the importance of applied data standards, alongside the critical role of methodology selection in future applications. Ensuring data quality improvement in the coming years should go together with careful consideration of methodologies. This workshop provided valuable insights for us.
Bayesian Federated Inference paper
Progress in medicine only happens when we can learn from data, but all too often data sets are too small and cannot be combined due to all kinds of practical and legal limitations. Especially for rare diseases (where data are always scarce) this is a serious obstacle. Therefore, we are delighted to announce the publication of our first paper on Bayesian Federated Inference (BFI) in collaboration with our partners at Radboud UMC! The BFI methodology allows one to harness the power of large shared data sets and infer patterns reliably without actually sharing any data.
Together, we can shape the future of data-driven medical innovation.
AISN Consortium Meeting
Saddle Point Science Europe, a technical partner in the AISN (Artificial Intelligence for Stroke Neurorehabilitation) consortium, works alongside esteemed companies, research institutes, and clinicians from across Europe. The consortium, dedicated to advancing stroke rehabilitation through Rehabilitation Gaming Systems (RGS), entrusted Saddle Point Science Europe with the development of mathematical prognostic and diagnostic models leveraging RGS parameters and data......
I4MDS General meeting:
Big data in Cancer Research
On February 9th and 10th the 2nd I4MDS General Meeting was held in The Hague, with support from the European Hematology Association. MDS, which stands for Myelodysplastic Neoplasm (see e.g. this recent review), is a disease of blood cells that is a specific precursor stage of acute myeloid leukemia......
