Thijs Smit ^1* Niels Aage ² Daniel Haschtmann ³ Stephen Ferguson ¹ Benedikt Helgason ¹

• ¹ ETH Zürich, Zurich, Switzerland
• ² Technical University of Denmark, Kongens Lyngby, Denmark
• ³ Schulthess Klinik, Zürich, Zürich, Switzerland

A full-scale topology optimization formulation has been developed to automate the design of cages used in instrumented transforaminal lumbar interbody fusion. The method incorporates the mechanical response of the adjacent bone structures in the optimization process, yielding patient-specific spinal fusion cages that both anatomically and mechanically conform to the patient, effectively mitigating subsidence risk when compared to generic, off-the-shelf cages and patient-specific devices. Insilico medical device testing on a cohort of 7 patients was performed to investigate the effectiveness of the anatomically and mechanically conforming devices using titanium and PEEK implant materials. A median reduction of the subsidence risk of 89% for titanium and 94% for PEEK implant materials was demonstrated compared to an offthe-shelf implant. A median reduction of 75% was achieved for PEEK implant material compared to an anatomically conforming implant. A credibility assessment of the computational model used to predict subsidence risk was provided according to the ASME V&V40-2018 standard. A full-scale topology optimization formulation to design anatomically and mechanically conforming patient-specific spinal fusion implants was tested, insilico, on a cohort of 7 patients. The anatomically and mechanically conforming patient-specific spinal fusion cages reduces the median subsidence risk by 89% for titanium and 94% for PEEK implant materials compared to an off-the-shelf implant. The method was similarly effective for patients with low and high bone quality. The credibility of the in-silico medical device testing procedure was evaluated according to the ASME V&V40-2018 standard.