Smarter surgery for children with skull disorders
Infant craniofacial abnormalities associated with premature skull bone fusion come under the umbrella of craniosynostosis(opens in new window). They affect one in 1 700 newborns and the resulting skull deformations can lead to severe health complications, including vision loss, neurodevelopmental delays and breathing difficulties. Traditional surgical approaches aim to reshape the skull and alleviate pressure. This usually involves the use of external or internal devices. However, predicting the outcomes of these surgeries has remained a challenge due to variability in bone properties and complex biomechanical interactions.
High-precision computational models
The ERC-funded CAD4FACE project set out to develop a computational framework for simulating children’s craniofacial skulls and guide surgical intervention on a patient-specific basis. A key objective was to understand the biomechanical properties of the skull bones in children born with craniosynostosis. To this end, the research team collected more than 300 bone samples during surgeries performed at Great Ormond Street Hospital(opens in new window) from over 250 children. These samples underwent high-resolution imaging and were mechanically tested to study the structural and material properties of bones from affected children. The generated data informed computational models that simulate skull reshaping after surgery and predict patient-specific surgical outcomes. Retrospective head growth data from cranioplasty surgeries were also included in model training. “By combining high-resolution imaging with biomechanical testing of paediatric skull samples, we’ve created models that can simulate post-operative skull shape with high accuracy,” explains principal investigator Silvia Schievano.
Project innovations
Several of CAD4FACE’s tools are now transitioning from research to clinical use. One notable breakthrough was the development of new craniofacial devices made from nitinol(opens in new window), a shape-memory alloy. These devices, designed and optimised through in silico testing, provide continuous, gentle force to reshape the skull, offering a less abrupt alternative to conventional steel solutions. The project also delivered a machine learning model trained on over 3 000 simulated virtual surgeries. This AI tool can predict surgical outcomes in real-time, providing a rapid decision-support system for clinicians. “We have taken the engineers and the computational burden out of the equation,” emphasises Schievano. “Surgeons can get immediate feedback on how different surgical options might impact head shape, enabling more confident planning.” Beyond simulations, CAD4FACE has pioneered the integration of virtual and augmented reality (VR/AR) in craniofacial surgery. These tools enable surgeons to plan procedures remotely, visualise predicted outcomes, and even project optimal surgical plans directly onto a patient’s head during operations. VR is also being used to improve communication with families, allowing them to visualise their child’s condition and expected outcomes in a more accessible way. For the future, says Schievano: "The integration of technologies such as computational modelling, machine learning, and extended realities is set to transform craniofacial surgery, making it more personalised, safer, and with better outcomes. This is particularly crucial for children born with craniofacial abnormalities, where precision and long-term impact are paramount."
Deployment and future prospects
Several project innovations have reached translational maturity and are ready for broader clinical translation. The VR/AR tools are deployed in teaching/training courses to support the adoption of personalised surgical planning across Europe. The new nitinol device is undergoing approval for first-in-child, while collaboration with industry partners is underway. Looking ahead, the team plans to expand the machine learning models, explore next-generation devices, and continue validating their tools through clinical trials. Through the CAD4FACE pipeline, they hope to personalise craniofacial surgery and improve clinical outcomes for children with craniosynostosis.
