ReviewPatient-specific bone modeling and analysis: The role of integration and automation in clinical adoption
Introduction
There is currently a general trend in healthcare for personalized medicine that includes use of imaging techniques, genetic analyses, and extensive biomarker evaluation to determine dedicated diagnosis or treatment protocols that differ for each specific patient. Such protocols appreciate the differences with respect to e.g. shape, morphology, genetics, and overall physiology of each individual in a much more extensive manner. As biomechanical engineers, we are interested in mechanical functioning of bodies, e.g. the shape and morphology of bone on the one hand and the structural performance related to its mechanical function on the other hand. The relationship between form and function is of primary importance since such types of biomechanical analysis could help us understand the evolutionary biomechanics of species as well as to study biological tissues with the structural and mechanical aspects as their major function.
This relationship brings us to Wolff׳s law, a topic that Rik Huiskes was so engaged with and many of Rik׳s PhD students were steered into this area. The topic certainly can contribute to orthopedics implant design or patient-specific modeling that might guide new diagnosis and/or treatment methods, something that Rik Huiskes pursued in his early career. During a later phase, he became more interested in the scientific issues themselves. As he phrased it so vividly: ‘If bone is the answer, then what is the question’ (Huiskes, 2000). Whereas the engineering approach followed by Julius Wolff more or less describes bone as an optimized material with maximal strength and minimal weight, Rik argued that there are no mathematical optimization rules for bone architecture. He claimed that bone follows only a biological regulatory process, producing a structure adapted to mechanical demands by the nature of its characteristics, adequate for evolutionary endurance. However, in biomechanical bone analysis, the mathematically described regulatory process could help us better understand the etiology of skeletal diseases and develop better techniques for their diagnosis and treatment.
Development of better techniques for diagnosis and treatment of skeletal diseases requires application of bone analysis techniques to specific patients. To do patient-specific bone analysis, one needs additional information pertaining to individual patients so as to personalize the biomechanical models of bone behavior, improve the sensitivity and specificity of the obtained analytical results, and provide the patients with personalized healthcare. That is why patient-specific bone analysis often starts from a set of images that are meant to provide the additional information needed for personalization of biomechanical models.
Images alone are, however, not enough for providing all the information that is required for patient-specific analysis of bones. At least two other types of information are needed, namely functional data and material properties. Functional data including the kinematics of patient׳s movements and the external forces exerted to the patient body during those movements (kinetic data) is needed to determine the internal musculoskeletal loads experienced by bones including joint reaction forces and muscle forces.
As for the material properties of bones, empirical relationships between bone density values measured using imaging modalities and elastic properties of bones are often used for obtaining patient-specific material properties. However, the empirical relationships are not available for all bones and may be drastically different from one patient to another (Eberle et al., 2013).
Several analytical tools (Fig. 1) have been developed to utilize the above-mentioned sources of information and establish the patient-specific relationship between bone shape, morphology, loading, and mechanical performance. The analytical tools may (1) process the image information and provide information regarding bone shape and morphology such as image processing algorithms and statistical models of shape and appearance, (2) use image information for determining the spatial distribution of mechanical properties such as material mapping relationships and material models, or (3) use functional data from musculoskeletal models for determining the loading conditions experienced by bones.
The purpose of this paper is not to review the patient-specific modeling technology, as this has been recently performed (Poelert et al., 2013). Instead, we focus on emerging tools, trends, and approaches that could revolutionize patient-specific modeling of bones and greatly facilitate the clinical acceptance of the technology.
If patient-specific bone analysis is going to be used in clinical settings, it needs to satisfy two criteria. First, it should provide clear added value for clinical diagnosis and treatment above that of currently available clinical tools. Second, it should be time and cost-effective. The first criterion relates to the accuracy of the analyses whereas the second relates to their cost-effectiveness. In this paper, we argue that integration and automation of the above-mentioned analytical tools could enable accurate and cost-effective patient-specific analysis of bones. Patient-specific finite element (FE) models are often used to intimately integrate the above-mentioned sources of information in analytical tools. In the following section, we introduce the different analytical tools and sources of information important for patient-specific bone analysis (Fig. 1). The different variants of every analytical tool or source of information will be discussed in the light of accuracy and cost. In the third section, we discuss possible ways of integrating and automating patient-specific bone analysis mainly in the context of patient-specific FE modeling. The paper concludes with a section discussing the limitations of currently available approaches and the required developments for clinical acceptance.
Section snippets
Components of patient-specific procedures
The different components used in patient-specific bone analyses (Fig. 1) are reviewed in this section. The different variants of every component are listed in Fig. 1. From left to right, the accuracy as well as the associated costs of the variants of different components increases. The cost is defined in a broad sense here, and includes a combination of monetary cost, time cost, and side effects such as the dosage of ionizing radiation. The variants that are currently seriously underdeveloped
Integration and automation
As previously mentioned, we argue here that clinical adoption of patient-specific bone analysis in general and patient-specific FE modeling in particular is dependent on two parameters: accuracy and cost-effectiveness of the available technology. Integration and automation are two important concepts that could help in improving both accuracy and cost-effectiveness of patient-specific bone analysis. By the term integration we mean combining the inputs and outputs of the components reviewed in
Discussion
Patient-specific bone modeling and analysis were discussed in the previous sections. It is clear that an accurate and cost-effective patient-specific modeling platform could be used for many diagnosis and treatments of several musculoskeletal pathologies. Important examples include design of optimal patient-specific implants and bone substitutes based on the possibilities of additive manufacturing technology that are well suited for the efficient production of a single implant with a dedicated
Conflict of interest
We hereby state that none of the authors has any financial or personal relationships with other people or organizations that could inappropriately influence (bias) our work.
References (104)
- et al.
On stabilization of loosened hip stems via cement injection into osteolytic cavities
Clin. Biomech.
(2012) - et al.
Comparison of an inhomogeneous orthotropic and isotropic material models used for FE analyses
Med. Eng. Phys.
(2008) - et al.
Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI
J. Magn. Reson.
(2011) Fabric tensor based boundary element analysis of porous solids
Eng. Anal. Bound. Elem.
(2011)- et al.
Prediction of strength and strain of the proximal femur by a CT-based finite element method
J. Biomech.
(2007) - et al.
Mechanical properties of open-cell metallic biomaterials manufactured using additive manufacturing
Mater. Des.
(2013) - et al.
Subject-specific modeling of the scapula bone tissue adaptation
J. Biomech.
(2013) Graphic-based musculoskeletal model for biomechanical analyses and animation
Med. Eng. Phys.
(2003)- et al.
Application of spherical and cylindrical wrapping algorithms in a musculoskeletal model of the upper limb
J. Biomech.
(2001) - et al.
Subject-specific bone loading estimation in the human distal radius
J. Biomech.
(2013)
Validation of a bone loading estimation algorithm for patient-specific bone remodelling simulations
J. Biomech.
Femoral strength is better predicted by finite element models than QCT and DXA
J. Biomech.
Force-plate based computation of ankle and hip strategies from double-inverted pendulum model
Clin. Biomech.
Anisotropic poroelasticity: fabric tensor formulation
Mech. Mater.
Human movement analysis using stereophotogrammetry: Part 4: assessment of anatomical landmark misplacement and its effects on joint kinematics
Gait Posture
Individual density–elasticity relationships improve accuracy of subject-specific finite element models of human femurs
J. Biomech.
A flexible software for tracking of markers used in human motion analysis
Comput. Methods Progr. Biomed.
Direct measurement of human movement by accelerometry
Med. Eng. Phys.
ESB clinical biomechanics award 2008: complete data of total knee replacement loading for level walking and stair climbing measured in vivo with a follow-up of 6–10 months
Clin. Biomech.
Mathematical relationships between bone density and mechanical properties: a literature review
Clin. Biomech.
A musculoskeletal model of the human lower extremity: the effect of muscle, tendon, and moment arm on the moment–angle relationship of musculotendon actuators at the hip, knee, and ankle
J. Biomech.
Comparison of the model-based and marker-based Roentgen stereophotogrammetry methods in a typical clinical setting
J. Arthroplast.
Instrumented implant for measuring tibiofemoral forces
J. Biomech.
The influence of footwear on knee joint loading during walking – in vivo load measurements with instrumented knee implants
J. Biomech.
Trabecular bone adaptation in a finite element frame model using load dependent fabric tensors
Mech. Mater.
Development of a fully automatic shape model matching (FASMM) system to derive statistical shape models from radiographs: application to the accurate capture and global representation of proximal femur shape
Osteoarthr. Cartil.
An EMG-driven musculoskeletal model to estimate muscle forces and knee joint moments in vivo
J. Biomech.
Full-field strain measurements in textile deformability studies
Compos. Part A: Appl. Sci. Manuf.
A review of 3D/2D registration methods for image-guided interventions
Med. Image Anal.
Toward a unifying theory of bone remodeling
Bone
A transformation method to estimate muscle attachments based on three bony landmarks
J. Biomech.
Discrimination of gender-, speed-, and shoe-dependent movement patterns in runners using full-body kinematics
Gait Posture
A symmetry invariant formulation of the relationship between the elasticity tensor and the fabric tensor
Mech. Mater.
A survey of advances in vision-based human motion capture and analysis
Comput. Vis. Image Underst.
Trabecular bone modulus–density relationships depend on anatomic site
J. Biomech.
Dependence of yield strain of human trabecular bone on anatomic site
J. Biomech.
An EMG-driven musculoskeletal model of the shoulder
Hum. Mov. Sci.
Comparison of isotropic and orthotropic material property assignments on femoral finite element models under two loading conditions
Med. Eng. Phys.
Estimation of distal radius failure load with micro-finite element analysis models based on three-dimensional peripheral quantitative computed tomography images
Bone
The relationship between two different mechanical cost functions and muscle oxygen consumption
J. Biomech.
Guided ultrasound wave propagation in intact and healing long bones
Ultrasound Med. Biol.
Predicting the subject-specific primary stability of cementless implants during pre-operative planning: preliminary validation of subject-specific finite-element models
J. Biomech.
Application of the digital volume correlation technique for the measurement of displacement and strain fields in bone: a literature review
J. Biomech.
An instrumented implant for vertebral body replacement that measures loads in the anterior spinal column
Med. Eng. Phys.
Roentgen stereophotogrammetry and metallic implants applied to patients with craniofacial anomalies
J. Biomech.
Micromovement of the tibial component in successful knee arthroplasty, studied by Roentgen stereophotogrammetry
J. Biomech.
Statistical shape and appearance models for fast and automated estimation of proximal femur fracture load using 2D finite element models
J. Biomech.
Statistical shape and appearance models of bones
Bone
Prediction of thoracic and lumbar vertebral body compressive strength: correlations with bone mineral density and vertebral region
Bone
Using digital image correlation to determine bone surface strains during loading and after adaptation of the mouse tibia
J. Biomech.
Cited by (39)
A combined active shape and mean appearance model for the reconstruction of segmental bone loss
2022, Medical Engineering and PhysicsCitation Excerpt :Integrating statistical models into the PSI design process can significantly reduce the specialist knowledge required by automating time-consuming activities such as segmentation and reconstruction, decreasing the cost and time needed for PSI design. Coupled with the advancement of additive manufacturing, statistical models make PSIs practically and financially more feasible for segmental bone repair [4,6,11,20]. The results found within this study motivates further investigation into automating segmental bone repair and PSI design procedures using the novel statistical model combination and methods discussed.
Benchmarking off-the-shelf statistical shape modeling tools in clinical applications
2022, Medical Image AnalysisFinite element analysis of the lumbar spine in adolescent idiopathic scoliosis subjected to different loads
2021, Computers in Biology and MedicineCitation Excerpt :FE analysis can analyze the mechanical changes in the pathological process of AIS, without any harm to the patient, and can provide strong support for clinical research. Research developments can be used for the prevention, and diagnosis of idiopathic spine in adolescents, assist in surgery and help in the design of curative effects and evaluation [55]. This study has certain limitations: There are many simplifications and assumptions in the FE model.
On the limits of finite element models created from (micro)CT datasets and used in studies of bone-implant-related biomechanical problems
2021, Journal of the Mechanical Behavior of Biomedical MaterialsCitation Excerpt :It must be noted that the rapid development of imaging technologies has also greatly contributed to the quality and range of computational simulations. These developments have been pivotal in creating personalized FEM models and for the fabrication of patient-specific implants (Zadpoor and Weinans, 2015). This novel approach is however still in its infancy and despite general optimism one should still bear in mind that computational simulations have limitations that should be recognized and understood.