Fully Automatic Segmentation of AP Pelvis X-rays via Random Forest Regression and Hierarchical Sparse Shape Composition

Knowledge of landmarks and contours in anteroposterior (AP) pelvis X-rays is invaluable for computer aided diagnosis, hip surgery planning and image-guided interventions. This paper presents a fully automatic and robust approach for landmarking and segmentation of both pelvis and femur in a conventional AP X-ray. Our approach is based on random forest regression and hierarchical sparse shape composition. Experiments conducted on 436 clinical AP pelvis x-rays show that our approach achieves an average point-to-curve error around 1.3 mm for femur and 2.2 mm for pelvis, both with success rates around 98%. Compared to existing methods, our approach exhibits better performance in both the robustness and the accuracy.

Making minimally invasive THR safe: conclusions from biomechanical simulation and analysis.

The use of smaller surgical incisions has become popularized for total hip arthroplasty (THR) because of the potential benefits of shorter recovery and improved cosmetic appearance. However, an increased incidence of serious complications has been reported. To minimize the risks of minimally invasive approaches to THR, we have developed an experimental approach which enables us to evaluate risk factors in these procedures through cadaveric simulations performed within the laboratory. During cadaveric hip replacement procedures performed via posterior and antero-lateral mini-incisions, pressures developed between the wound edges and the retractors were approximately double those recorded during conventional hip replacement using Charnley retractors (p < 0.01). In MIS procedures performed via the dual-incision approach, lack of direct visualisation of the proximal femur led to misalignment of broaches and implants with increased risk of cortical fracture during canal preparation and implant insertion. Cadaveric simulation of surgical procedures allows surgeons to measure variables affecting the technical success of surgery and to master new procedures without placing patients at risk.

Effects of enamel matrix proteins in combination with a bovine-derived natural bone mineral for the repair of bone defects.

OBJECTIVES Previously, the use of enamel matrix derivative (EMD) in combination with a natural bone mineral (NBM) was able to stimulate periodontal ligament cell and osteoblast proliferation and differentiation. Despite widespread use of EMD for periodontal applications, the effects of EMD on bone regeneration are not well understood. The aim of the present study was to test the ability of EMD on bone regeneration in a rat femur defect model in combination with NBM. MATERIALS AND METHODS Twenty-seven rats were treated with either NBM or NBM + EMD and assigned to histological analysis at 2, 4, and 8 weeks. Defect morphology and mineralized bone were assessed by μCT. For descriptive histology, hematoxylin and eosin staining and Safranin O staining were performed. RESULTS Significantly more newly formed trabecular bone was observed at 4 weeks around the NBM particles precoated with EMD when compared with NBM particles alone. The drilled control group, in contrast, achieved minimal bone regeneration at all three time points (P < 0.05). CONCLUSIONS The present results may suggest that EMD has the ability to enhance the speed of new bone formation when combined with NBM particles in rat osseous defects. CLINICAL RELEVANCE These findings may provide additional clinical support for the combination of EMD with bone graft for the repair of osseous and periodontal intrabony defects.

Bone Volume Fraction and Fabric Anisotropy Are Better Determinants of Trabecular Bone Stiffness than Other Morphological Variables

As our population ages, more individuals suffer from osteoporosis. This disease leads to impaired trabecular architecture and increased fracture risk. It is essential to understand how morphological and mechanical properties of the cancellous bone are related. Morphologyelasticity relationships based on bone volume fraction (BV/TV) and fabric anisotropy explain up to 98% of the variation in elastic properties. Yet, other morphological variables such as individual trabeculae segmentation (ITS) and trabecular bone score (TBS) could improve the stiffness predictions. A total of 743 micro-computed tomography reconstructions of cubic trabecular bone samples extracted from femur, radius, vertebrae and iliac crest were analysed. Their morphology was assessed via 25 variables and their stiffness tensor (inline image) was computed from six independent load cases using micro finite element analyses. Variance inflation factors were calculated to evaluate collinearity between morphological variables and decide upon their inclusion in morphology-elasticity relationships. The statistically admissible morphological variables were included in a multi-linear regression modelling the dependent variable inline image. The contribution of each independent variable was evaluated (ANOVA). Our results show that BV/TV is the best determinant of inline image (inline image=0.889), especially in combination with fabric (inline image=0.968). Including the other independent predictors hardly affected the amount of variance explained by the model (inline image=0.975). Across all anatomical sites, BV/TV explained 87% of the variance of the bone elastic properties. Fabric further described 10% of the bone stiffness, but the improvement in variance explanation by adding other independent factors was marginal (<1%). These findings confirm that BV/TV and fabric are the best determinants of trabecular bone stiffness and show, against common belief, that other morphological variables do not bring any further contribution. These overall conclusions remain to be confirmed for specific bone diseases and post-elastic properties.

Intraoperative Mechanical Measurement of Bone Quality With the DensiProbe.

Reduced bone stock can result in fractures that mostly occur in the spine, distal radius, and proximal femur. In case of operative treatment, osteoporosis is associated with an increased failure rate. To estimate implant anchorage, mechanical methods seem to be promising to measure bone strength intraoperatively. It has been shown that the mechanical peak torque correlates with the local bone mineral density and screw failure load in hip, hindfoot, humerus, and spine in vitro. One device to measure mechanical peak torque is the DensiProbe (AO Research Institute, Davos, Switzerland). The device has shown its effectiveness in mechanical peak torque measurement in mechanical testing setups for the use in hip, hindfoot, and spine. In all studies, the correlation of mechanical torque measurement and local bone mineral density and screw failure load could be shown. It allows the surgeon to judge local bone strength intraoperatively directly at the region of interest and gives valuable information if additional augmentation is needed. We summarize methods of this new technique, its advantages and limitations, and give an overview of actual and possible future applications.

An epidemiological evaluation of pediatric long bone fractures - a retrospective cohort study of 2716 patients from two Swiss tertiary pediatric hospitals.

BACKGROUND Children and adolescents are at high risk of sustaining fractures during growth. Therefore, epidemiological assessment is crucial for fracture prevention. The AO Comprehensive Injury Automatic Classifier (AO COIAC) was used to evaluate epidemiological data of pediatric long bone fractures in a large cohort. METHODS Data from children and adolescents with long bone fractures sustained between 2009 and 2011, treated at either of two tertiary pediatric surgery hospitals in Switzerland, were retrospectively collected. Fractures were classified according to the AO Pediatric Comprehensive Classification of Long Bone Fractures (PCCF). RESULTS For a total of 2716 patients (60% boys), 2807 accidents with 2840 long bone fractures (59% radius/ulna; 21% humerus; 15% tibia/fibula; 5% femur) were documented. Children's mean age (SD) was 8.2 (4.0) years (6% infants; 26% preschool children; 40% school children; 28% adolescents). Adolescent boys sustained more fractures than girls (p < 0.001). The leading cause of fractures was falls (27%), followed by accidents occurring during leisure activities (25%), at home (14%), on playgrounds (11%), and traffic (11%) and school accidents (8%). There was boy predominance for all accident types except for playground and at home accidents. The distribution of accident types differed according to age classes (p < 0.001). Twenty-six percent of patients were classed as overweight or obese - higher than data published by the WHO for the corresponding ages - with a higher proportion of overweight and obese boys than in the Swiss population (p < 0.0001). CONCLUSION Overall, differences in the fracture distribution were sex and age related. Overweight and obese patients seemed to be at increased risk of sustaining fractures. Our data give valuable input into future development of prevention strategies. The AO PCCF proved to be useful in epidemiological reporting and analysis of pediatric long bone fractures.

Fractures in Children and Adolescents

Fractures of the growing bone require fixation techniques, which preclude any injury to the growth plate regions. This requirement is met by Elastic Stable Intramedullary Nails (ESIN) which are positioned between both metaphyseal regions. Pronounced malposition and/or shortening, open fractures and fractures with impending skin perforation are indications for clavicle nailing in adolescents. Retrograde nailing with two elastic nails, inserted from lateral, is the method of choice for stabilization of humerus fractures. In radial neck fractures with severe tilting of the radial head, a retrograde nail may reduce and fix the head. In Monteggia lesions, the ulna fracture is reduced and fixed with an antegrade nail. Forearm fractures with unacceptable axial deviation are reduced and fixed with one antegrade nail in the ulna and a retrograde nail in the radius. Ascending elastic nailing is done for femur shaft and proximal femur fractures. The medial and lateral entry sites are located above the distal physis. End caps are used to prevent shortening in spiral and multiple segment fractures. Fractures of the distal third of the femur are nailed in a descending technique. The entry sites of two nails are located on the lateral cortex below the greater trochanter. Combined tibia and fibula fractures, open fractures and unstable fracture types such as spiral and multifragmental tibia fractures are good indications for ESIN. Descending nailing is the method of choice. The nail entry points are medially and laterally distal to the apophysis of the proximal tibia. Thorough knowledge of each fracture type, fracture location and age specific healing pattern is necessary for safe and effective treatment of pediatric fractures

Möller-Barlow disease in archaeology: Preliminary study of biochemical detection

Human bone is the most direct source for reconstructing health and living conditions of ancient populations. However, many diseases remain undetected in palaeopathology. Möller-Barlow disease (scurvy) is a historically well-documented metabolic disease and must have been common in clinical and sub-clinical severity. Due to long incubation periods and the subtle nature of bone changes osteological evidence is relatively rare (Brickley & Ives 2008). Möller-Barlow disease is caused by deficiency of dietary vitamin C (ascorbic acid) and evokes symptoms like fatigue, haemorrhage, inflammations, delayed wound healing and pain. Vitamin C is a cofactor for the hydroxylation of the amino acids proline and lysine which are essential for the production of intact connective tissue by cross-linking the propeptides in collagen. In a preliminary study we tested the detectability of Möller-Barlow disease by analysis of relative quantitative variability of hydroxylated amino acids in collagen (Pendery & Koon 2013). Samples (N=9) were taken from children with (n=3, cranium, femur, tibia) and without (n=4, cranium, femur, tibia) apparent bone reactions indicative of Möller-Barlow disease, as well as from adults with lethal traumata (n=2; negative controls). The skeletal remains originated from two early medieval cemeteries from Switzerland. Gas chromatographic (GC) analysis revealed minor differences between the samples. So far children with no pathologic alterations had fairly same values as negative controls while children with bone reactions paradoxically exhibited even slightly higher values of hydroxyproline and hydroxylysine. Future research demands for larger sample size and has to discuss sampling strategies. Beside possible misdiagnosis of Möller-Barlow disease it is arguable if only the newly built bone should be analysed even though this could lead to problems related to small sample quantity. It also remains to be seen to which extent varying turnover rates of different skeletal elements, especially in children, must be taken into account.

Fully Automatic CT Segmentation for Computer-Assisted Pre-operative Planning of Hip Arthroscopy

Extraction of both pelvic and femoral surface models of a hip joint from CT data for computer-assisted pre-operative planning of hip arthroscopy is addressed. We present a method for a fully automatic image segmentation of a hip joint. Our method works by combining fast random forest (RF) regression based landmark detection, atlas-based segmentation, with articulated statistical shape model (aSSM) based hip joint reconstruction. The two fundamental contributions of our method are: (1) An improved fast Gaussian transform (IFGT) is used within the RF regression framework for a fast and accurate landmark detection, which then allows for a fully automatic initialization of the atlas-based segmentation; and (2) aSSM based fitting is used to preserve hip joint structure and to avoid penetration between the pelvic and femoral models. Validation on 30 hip CT images show that our method achieves high performance in segmenting pelvis, left proximal femur, and right proximal femur surfaces with an average accuracy of 0.59 mm, 0.62 mm, and 0.58 mm, respectively.

Articulated Statistical Shape Model-Based 2D-3D Reconstruction of a Hip Joint

In this paper, reconstruction of three-dimensional (3D) patient-specific models of a hip joint from two-dimensional (2D) calibrated X-ray images is addressed. Existing 2D-3D reconstruction techniques usually reconstruct a patient-specific model of a single anatomical structure without considering the relationship to its neighboring structures. Thus, when those techniques would be applied to reconstruction of patient-specific models of a hip joint, the reconstructed models may penetrate each other due to narrowness of the hip joint space and hence do not represent a true hip joint of the patient. To address this problem we propose a novel 2D-3D reconstruction framework using an articulated statistical shape model (aSSM). Different from previous work on constructing an aSSM, where the joint posture is modeled as articulation in a training set via statistical analysis, here it is modeled as a parametrized rotation of the femur around the joint center. The exact rotation of the hip joint as well as the patient-specific models of the joint structures, i.e., the proximal femur and the pelvis, are then estimated by optimally fitting the aSSM to a limited number of calibrated X-ray images. Taking models segmented from CT data as the ground truth, we conducted validation experiments on both plastic and cadaveric bones. Qualitatively, the experimental results demonstrated that the proposed 2D-3D reconstruction framework preserved the hip joint structure and no model penetration was found. Quantitatively, average reconstruction errors of 1.9 mm and 1.1 mm were found for the pelvis and the proximal femur, respectively.

Fully automatic segmentation of AP pelvis X-rays via random forest regression with efficient feature selection and hierarchical sparse shape composition

In clinical practice, traditional X-ray radiography is widely used, and knowledge of landmarks and contours in anteroposterior (AP) pelvis X-rays is invaluable for computer aided diagnosis, hip surgery planning and image-guided interventions. This paper presents a fully automatic approach for landmark detection and shape segmentation of both pelvis and femur in conventional AP X-ray images. Our approach is based on the framework of landmark detection via Random Forest (RF) regression and shape regularization via hierarchical sparse shape composition. We propose a visual feature FL-HoG (Flexible- Level Histogram of Oriented Gradients) and a feature selection algorithm based on trace radio optimization to improve the robustness and the efficacy of RF-based landmark detection. The landmark detection result is then used in a hierarchical sparse shape composition framework for shape regularization. Finally, the extracted shape contour is fine-tuned by a post-processing step based on low level image features. The experimental results demonstrate that our feature selection algorithm reduces the feature dimension in a factor of 40 and improves both training and test efficiency. Further experiments conducted on 436 clinical AP pelvis X-rays show that our approach achieves an average point-to-curve error around 1.2 mm for femur and 1.9 mm for pelvis.

Automatic X-ray landmark detection and shape segmentation via data-driven joint estimation of image displacements

In this paper, we propose a new method for fully-automatic landmark detection and shape segmentation in X-ray images. To detect landmarks, we estimate the displacements from some randomly sampled image patches to the (unknown) landmark positions, and then we integrate these predictions via a voting scheme. Our key contribution is a new algorithm for estimating these displacements. Different from other methods where each image patch independently predicts its displacement, we jointly estimate the displacements from all patches together in a data driven way, by considering not only the training data but also geometric constraints on the test image. The displacements estimation is formulated as a convex optimization problem that can be solved efficiently. Finally, we use the sparse shape composition model as the a priori information to regularize the landmark positions and thus generate the segmented shape contour. We validate our method on X-ray image datasets of three different anatomical structures: complete femur, proximal femur and pelvis. Experiments show that our method is accurate and robust in landmark detection, and, combined with the shape model, gives a better or comparable performance in shape segmentation compared to state-of-the art methods. Finally, a preliminary study using CT data shows the extensibility of our method to 3D data.

FACTS: Fully Automatic CT Segmentation of a Hip Joint

Extraction of surface models of a hip joint from CT data is a pre-requisite step for computer assisted diagnosis and planning (CADP) of periacetabular osteotomy (PAO). Most of existing CADP systems are based on manual segmentation, which is time-consuming and hard to achieve reproducible results. In this paper, we present a Fully Automatic CT Segmentation (FACTS) approach to simultaneously extract both pelvic and femoral models. Our approach works by combining fast random forest (RF) regression based landmark detection, multi-atlas based segmentation, with articulated statistical shape model (aSSM) based fitting. The two fundamental contributions of our approach are: (1) an improved fast Gaussian transform (IFGT) is used within the RF regression framework for a fast and accurate landmark detection, which then allows for a fully automatic initialization of the multi-atlas based segmentation; and (2) aSSM based fitting is used to preserve hip joint structure and to avoid penetration between the pelvic and femoral models. Taking manual segmentation as the ground truth, we evaluated the present approach on 30 hip CT images (60 hips) with a 6-fold cross validation. When the present approach was compared to manual segmentation, a mean segmentation accuracy of 0.40, 0.36, and 0.36 mm was found for the pelvis, the left proximal femur, and the right proximal femur, respectively. When the models derived from both segmentations were used to compute the PAO diagnosis parameters, a difference of 2.0 ± 1.5°, 2.1 ± 1.6°, and 3.5 ± 2.3% were found for anteversion, inclination, and acetabular coverage, respectively. The achieved accuracy is regarded as clinically accurate enough for our target applications.

Complementary passive hydrotherapy WATSU (WaterShiatsu) in the rehabilitation of a severely traumatized accident survivor – A prospective case report

Introduction: The implementation of complementary and alternative therapies into conventional treatment schemes is gaining popularity. However, their use is widely depending on patients’ drive. This case-report focuses on a patient’s experience of the integration of WATSU (WaterShiatsu) in rehabilitative care. Methods: Patient: A 52 year old woman survived a severe motorcycle-accident in which she sustained several fractures on the right side of her body, including ribs, pelvis, and femur. After discharge from stationary care, she independently added WATSU to her rehabilitative regimen. Treatment approach: WATSU is a passive form of hydrotherapy in warm water that aims at relaxation, pain relief, and a sense of secureness. In the reported case, an experienced WATSU-therapist who is also trained in physiotherapy and psychosomatics delivered weekly sessions of one hour duration. Measures used: Qualitative data were collected by patient’s diary. Also the therapist’s notes including The Patient Specific Functional Scale (PSFS) were considered. Results: The patient associated WATSU with trunk mobilization (followed by ameliorated breath), reconciliation with her body, and emotional discharge. She ascribed WATSU lasting effects on her body-image. The therapist employed WATSU for careful mobilization and to equalize awareness throughout the body. The PSFS displayed continuous improvement in all categories except usage of public transportation. Due to complications (elevated inflammation markers) only 6 of 8 scheduled sessions were administered. Conclusions: WATSU was experienced helpful in approaching conditions that are difficult to address by conventional physiotherapy. In early rehabilitation, additional medical/physiotherapeutic skills of contributing complementary therapists are advocated.

Nonlinear quasi-static finite element simulations predict in vitro strength of human proximal femora assessed in a dynamic sideways fall setup

Osteoporotic proximal femur fractures are caused by low energy trauma, typically when falling on the hip from standing height. Finite element simulations, widely used to predict the fracture load of femora in fall, usually include neither mass-related inertial effects, nor the viscous part of bone's material behavior. The aim of this study was to elucidate if quasi-static non-linear homogenized finite element analyses can predict in vitro mechanical properties of proximal femora assessed in dynamic drop tower experiments. The case-specific numerical models of thirteen femora predicted the strength (R2=0.84, SEE=540 N, 16.2%), stiffness (R2=0.82, SEE=233 N/mm, 18.0%) and fracture energy (R2=0.72, SEE=3.85 J, 39.6%); and provided fair qualitative matches with the fracture patterns. The influence of material anisotropy was negligible for all predictions. These results suggest that quasi-static homogenized finite element analysis may be used to predict mechanical properties of proximal femora in the dynamic sideways fall situation.