Publicação
Development of procedures for the design, optimization and manufacturing of customized orthopaedic and trauma implants: Geometrical/anatomical modelling from 3D medical imaging
| Resumo: | The introduction of imaging techniques in 1970 is one of the most relevant historical milestones in modern medicine. Medical imaging techniques have dramatically changed our understanding of the Human anatomy and physiology. The ability to non-invasively extract visual information allowed, not only the three-dimensional representation of the internal organs and musculo-skeletal system, but also the simulation of surgical procedures, the execution of computer aided surgeries, the development of more accurate biomechanical models, the development of custom-made implants, among others. The combination of the most advanced medical imaging systems with the most advanced CAD and CAM techniques, may allow the development of custom-made implants that meet patient-speci c traits. The geometrical and functional optimization of these devices may increase implant life-expectancy, especially in patients with marked deviations from the anatomical standards. In the implant customization protocol from medical image data, there are several steps that need to be followed in a sequential way, namely: Medical Image Processing and Recovering; Accurate Image Segmentation and 3D Surface Model Generation; Geometrical Customization based on CAD and CAE techniques; FEA Optimization of the Implant Geometry; and Manufacturing using CAD-CAM Technologies. This work aims to develop the necessary procedures for custom implant development from medical image data. This includes the extraction of highly accurate three-dimensional representation of the musculo-skeletal system from the Computed Tomography imaging, and the development of customized implants, given the speci c requirements of the target anatomy, and the applicable best practices found in the literature. A two-step segmentation protocol is proposed. In the rst step the region of interest is pre-segmented in order to obtain a good approximation to the desired geometry. Next, a fully automatic segmentation re nement is applied to obtain a more accurate representation of the target domain. The re nement step is composed by several sub-steps, more precisely, the recovery of the original image, considering the limiting resolution of the imaging system; image cropping; image interpolation; and segmentation re nement over the up-sampled domain. Highly accurate segmentations of the target domain were obtained with the proposed pipeline. The limiting factor to the accurate description of the domain accuracy is the image acquisition process, rather the following image processing, segmentation and surface meshing steps. The new segmentation pipeline was used in the development of three tailor-made implants, namely, a tibial nailing system, a mandibular implant, and a Total Hip Replacement system. Implants optimization is carried with Finite Element Analysis, considering the critical loading conditions that may be applied to each implant in working conditions. The new tibial nailing system is able of sustaining critical loads without implant failure; the new mandibular endoprosthesis that allows the recovery of the natural stress and strain elds observed in intact mandibles; and the Total Hip Replacement system that showed comparable strain shielding levels as commercially available stems. In summary, in the present thesis the necessary procedures for custom implant design are investigated, and new algorithms proposed. The guidelines for the characterization of the image acquisition, image processing, image segmentation and 3D reconstruction are presented and discussed. This new image processing pipeline is applied and validated in the development of the three abovementioned customized implants, for di erent medical applications and that satisfy speci c anatomical needs. |
|---|---|
| Autores principais: | Pinheiro, Manuel da Silva |
| Assunto: | Ciências Médicas::Biotecnologia Médica |
| Ano: | 2015 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | The introduction of imaging techniques in 1970 is one of the most relevant historical milestones in modern medicine. Medical imaging techniques have dramatically changed our understanding of the Human anatomy and physiology. The ability to non-invasively extract visual information allowed, not only the three-dimensional representation of the internal organs and musculo-skeletal system, but also the simulation of surgical procedures, the execution of computer aided surgeries, the development of more accurate biomechanical models, the development of custom-made implants, among others. The combination of the most advanced medical imaging systems with the most advanced CAD and CAM techniques, may allow the development of custom-made implants that meet patient-speci c traits. The geometrical and functional optimization of these devices may increase implant life-expectancy, especially in patients with marked deviations from the anatomical standards. In the implant customization protocol from medical image data, there are several steps that need to be followed in a sequential way, namely: Medical Image Processing and Recovering; Accurate Image Segmentation and 3D Surface Model Generation; Geometrical Customization based on CAD and CAE techniques; FEA Optimization of the Implant Geometry; and Manufacturing using CAD-CAM Technologies. This work aims to develop the necessary procedures for custom implant development from medical image data. This includes the extraction of highly accurate three-dimensional representation of the musculo-skeletal system from the Computed Tomography imaging, and the development of customized implants, given the speci c requirements of the target anatomy, and the applicable best practices found in the literature. A two-step segmentation protocol is proposed. In the rst step the region of interest is pre-segmented in order to obtain a good approximation to the desired geometry. Next, a fully automatic segmentation re nement is applied to obtain a more accurate representation of the target domain. The re nement step is composed by several sub-steps, more precisely, the recovery of the original image, considering the limiting resolution of the imaging system; image cropping; image interpolation; and segmentation re nement over the up-sampled domain. Highly accurate segmentations of the target domain were obtained with the proposed pipeline. The limiting factor to the accurate description of the domain accuracy is the image acquisition process, rather the following image processing, segmentation and surface meshing steps. The new segmentation pipeline was used in the development of three tailor-made implants, namely, a tibial nailing system, a mandibular implant, and a Total Hip Replacement system. Implants optimization is carried with Finite Element Analysis, considering the critical loading conditions that may be applied to each implant in working conditions. The new tibial nailing system is able of sustaining critical loads without implant failure; the new mandibular endoprosthesis that allows the recovery of the natural stress and strain elds observed in intact mandibles; and the Total Hip Replacement system that showed comparable strain shielding levels as commercially available stems. In summary, in the present thesis the necessary procedures for custom implant design are investigated, and new algorithms proposed. The guidelines for the characterization of the image acquisition, image processing, image segmentation and 3D reconstruction are presented and discussed. This new image processing pipeline is applied and validated in the development of the three abovementioned customized implants, for di erent medical applications and that satisfy speci c anatomical needs. |
|---|