Artikel
Patient specific implants for skull defects: universal algorithm for conceptual design
Suche in Medline nach
Autoren
-
Daniel Schoeni
- Universitätsklinikum Tübingen, Klinik für Neurochirurgie, Bereich Pädiatrische Neurochirurgie, Tübingen, Deutschland, Department of Neurosurgery, Inselspital, University Hospital Bern, and University of Bern, Bern, Switzerland, Bern, Switzerland
-
Andreas Nowacki
- Department of Neurosurgery, Inselspital, University Hospital Bern, and University of Bern, Bern, Switzerland, Bern, Switzerland
-
Andreas Raabe
- Inselspital, Bern University Hospital, Department of Neurosurgery, Bern, Switzerland
-
Philippe Schucht
- Department of Neurosurgery, Inselspital, University Hospital Bern, and University of Bern, Bern, Switzerland, Bern, Switzerland
| Veröffentlicht: | 9. Juni 2017 |
|---|
Gliederung
Text
Objective: Cranioplasty after decompressive hemicraniectomy or large format craniotomies is a challenging procedure especially in cases when a patient's own cranial flap is unavailable. Recent advances in computer-aided design and 3-dimensional (3-D) printing have enabled fabrication of patient specific implants (PSI) from a variety of polymer, ceramic, or metal components. Good functional as well as satisfying cosmetic postoperative results have been reported. However, computer-aided design and manufacturing of PSI requires fundamental knowledge in 3-D-processing software and can be difficult and time-consuming in a variety of cases. We aim at presenting first results of a fully-automated computer-based algorithm for PSI design (CAPSID) requiring no manual operation of 3-D-processing software by the surgeon.
Methods: Based on the patient`s defect skull computed tomography (CT) scan we developed an algorithm implementing about 20 computational steps including rigid-, non-rigid fusion algorithms, mirror function, fusion and subtraction functions to calculate a 3-D mould (CAPSID). The algorithm works fully-automated requiring no user´s manual adaptations. CAPSID was applied retrospectively in 9 patients with different types and sizes of skull defects (left and right decompressive hemicraniectomy, craniotomies crossing the midline, skull defects containing a subgaleal drain and external ventricular drain) to test for correct feasibility of the algorithm.
Results: Application of CAPSID led to successful construction of the 3-D mould in every 9 patients. Mean running time of the algorithm was 09:32 0.45 min. The resulting 3-D mould was constructed with high accuracy in 7 of 9 patients. Acquisition of CT scans with 3 mm slice thickness and a subgaleal drain crossing the skull defect affected accurate 3-D-mould construction.
Conclusion: Application of CAPSID for conception of 3-D moulds for manufacturing of PSI seems feasible in every case. Improvement of the algorithm to achieve high accuracy is needed in cases of CT scan acquisition with thick slices and subgaleal drains crossing the skull defect. A prospective cohort study is currently under investigation to confirm our results and test clinical practicability and patient satisfaction (NCT02828306).
