The objective of our study is to evaluate the accuracy of an X-ray based image segmentation system for patient specific instrument (PSI) design or any other surgical application that requires 3D modeling of the knee. The process requires two bilateral short film X-ray images of knee and a standing long film image of the leg including the hip and ankle. The short film images are acquired with an X-ray positioner device that is embedded with fiducial markers to correct for setup variation in source and cassette position. An automated image segmentation algorithm, based on a statistical model that couples knee bone shape and radiographic appearance, calculates 3D surface models of the knee from the bi-lateral short films (Imorphics, Manchester UK) (Figure 1). Surface silhouettes are used to inspect and refine the automatically generated segmentation; the femur and tibia mechanical axes are then calculated using automatically generated surface model landmarks combined with user-defined markups of the hip and ankle center from the standing long film (Figure 2). The accuracy of the 2D/3D segmentation system was evaluated using simulated X-ray imagery generated from one-hundred osteoarthritic, lower limb CT image samples using the Insight Toolkit (Kitware, Inc.). Random, normally distributed variations in source and cassette positions were included in the dataset. Surface accuracy was measured using root-mean-square (RMS) point-to-surface (P2S) distance calculations with respect to paired benchmark CT segmentations. Landmark accuracy was calculated by measuring angular differences between the 2D/3D generated femur and tibia mechanical tibia with respect to paired CT-generated landmark data. The paired RMS sample mean and standard deviation of femur P2S errors on the distal quarter of the femur after auto-segmentation was 1.08±0.20mm. The RMS sample mean and standard deviation of tibia P2S errors on the proximal quarter of the tibia after auto-segmentation was 1.16±0.25mm. The paired sample mean and standard deviation of the femur and tibia mechanical axis accuracy with respect to benchmark CT data landmarks were 0.02±0.42[deg] and −0.33±0.56[deg], respectively. Per surface-vertex sample RMS P2S errors are illustrated in Figure 3. Visual inspection of RMS results found the automatically segmented femur to be very accurate in the shaft, distal condyles, and posterior condyles, which are important for PSI guide fit and accurate planning. Similarly, the automatically segmented tibia was very accurate in the shaft and plateaus, which are also important for PSI guide fit. Osteophytes resulted in some RMS differences (Figure 3), as was expected due to the know limitations of osteophyte imaging with X-ray. PSI-type applications that utilize X-ray should account for osteophyte segmentation error. Overall, our results based on simulated radiographic data demonstrate that X-ray based 2D/3D segmentation is a viable tool for use in orthopaedic applications that require accurate 3D segmentations of knee bones.
Leg length and offset discrepancy resulting from Total Hip Replacement (THR) is a major cause of concern for the orthopaedic community. The inability to substitute the proximal portion of the native femur with a device that suitably mimics the pre-operative offset and head height can lead to loss of abductor power, instability, lower back pain and the need for orthodoses (1). Contemporary devices are manufactured based on predicate studies (2–4) to cater for the variations within the patient demographic. Stem variants, modular necks and heads are often provided to meet this requirement. The number of components and instruments that manufacturers are prepared to supply however is limited by cost and an unwillingness to introduce unnecessary complexity. This can restrict their ability to achieve the pre-osteoarthritic head centre for all patient morphologies. Corin has developed bone conserving prosthesis (MiniHip™) to better replicate the physiological load distribution in the femur. This study assesses whether the MiniHip™ prosthesis can better match the pre-osteoarthritic head centre for patient demographics when compared to contemporary long stem devices. The Dorr classification is a well accepted clinical method for defining femoral endosteal morphology (5). This is often used by the surgeon to select the appropriate type and size of stem for the individual patient. It is accepted that a strong correlation exists between Flare Index (FI), characterising the thinning of cortical walls and development of ‘stove-pipe’ morphology, and age, in particular for females (Table 1) (3). A statistical model of the proximal femur was built from 30 full length femoral scans (Imorphics, UK). Minimum and maximum intramedullary measurements calculated from the statistical model were applied to relationships produced by combining Corins work with that of prior authors (Table 2) (2; 3; 6). This data was then used to generate 2D CAD models into which implants were inserted to compare the head centres achievable with a MiniHip™ device compared to those of a contemporary long stem.Introduction:
Method:
Bone resorption around hip stems, in particular periprosthetic bone loss, is a common observation post-operatively. A number of factors influence the amount of bone loss over time and the mechanical environment following total hip replacement (THR) is important; conventional long stem prostheses have been shown to transfer loads distally, resulting in bone loss of the proximal femur. More conservative, short stems have been recently introduced to attempt to better replicate the physiological load distribution in the femur. The aim of this study was to evaluate the bone mineral density (BMD) change over time, in a femur implanted with either a short or a long stem. Finite element models of two implants, a short (Minihip, Corin, UK) and long (Metafix, Corin, UK) hip stem were used to simulate bone remodeling under a physiological load condition (stair climbing). The magnitudes and directions of the muscle forces and joint reaction force were obtained from Heller et al (2001, 2005). An unimplanted femur was also simulated. A strain-adaptive remodelling theory (Scannel & Prendergast 2009) was utilised to simulate remodelling in the bone after virtual implantation. COMSOL Multiphysics software was used for the analysis. The strain component of the remodelling stimulus was strain energy density per unit mass. This was calculated in the continuum model from the strain energy density, and apparent density. Bone mass was adapted using a site-specific approach in an attempt to return the local remodelling stimulus to the equilibrium stimulus level (calculated from the unimplanted femur). The minimal inhibitory signal proposed by Frost (1964), was included in the model and described by a ‘lazy zone’, where no bone remodelling occurred. The three dimensional geometry of the femur was constructed from computed tomography data of the donor (female, 44 years old, right side). Elemental bone properties were assigned from the Hounsfield Unit values of the CT scans. The elastic modulus of the bone was assumed to be isotropic and was determined using a relationship to the apparent bone density (Frost 1964, Rho 1995). The Poisson's ratio for the bone regions varied between 0.2 and 0.32 depending on the apparent density of the bone (Stulpner 1997). The period of implantation analysed was 2 years. The muscle forces and joint contact loads applied were ramped linearly from zero to full load over a period of two weeks, representing the estimated post operative rest period of a patient.INTRODUCTION
METHODS
