A digital twin roadmap for adult spinal deformity.

Abstract

Adult spinal deformity (ASD) surgery is a high-risk, high-cost domain in which preoperative planning plays a critical role in determining alignment goals, surgical strategy, and the balance between deformity correction and treatment-related complications, yet decisions still rely heavily on static radiographs, classification systems, and surgeon experience. In parallel, medical digital twins, continuously updated, patient-specific computational models that simulate alternative treatment paths, are emerging in other medical disciplines but have not been formally articulated for ASD. In this narrative review, we adapt the recently proposed five-component medical digital twin framework (patient, data connection, patient-in-silico, interface, and synchronization policy) to adult spinal deformity by mapping each component to ASD-specific data sources, computational models, clinical workflows, and validation pathways. We outline how existing tools in ASD, including classification systems, frailty and bone-health metrics, biomechanical modeling, motion analysis, mobile health data, and machine learning risk calculators, can serve as inputs to an ASD digital twin. We highlight priority use cases, such as choosing the magnitude of surgery, bone-health-aware planning, pelvic fixation and rod strategy, and selection among open, hybrid, and minimally invasive approaches, all reframed as comparative simulations with explicit uncertainty bands. Finally, we propose a staged roadmap, from minimal viable twins evaluated via retrospective replay, to shadow-mode deployment, to oversighted clinical pilots, while detailing challenges in data harmonization, model validity, governance, and equity. Rather than claiming that a mature ASD digital twin exists today, this work offers a practical blueprint for how digital twins could evolve into auditable, patient-specific decision-support tools that augment, rather than replace, surgeon judgment in spinal deformity surgery.