ONFH and finite element model There were many ONFH finite element models exiting, such as cone model, spherical model and pathological models. However, with further development of the technology and the needs from clinical practices, model design and building experiments were required to be invented on the basis of previous modeling application by using the original CT data from ONFH patients[12]. Li et al.[13] used the 142-layer of 0.625-mm-thick consecutive CT Dicom format images to scan the ONFH, and then define the threshold of bone tissue, extract each outline, partition each edges, edit and repair by hole processing. Finally, they created the three-dimensional geometry model and meshed body line, assigned, and generate a model. The author believes that this method can quickly establish realistic femoral head in computer and calculate the precise three-dimensional finite element model of ONFH. Undeniably, some limitations exist in modeling by CT, due to its poor ability to detect cartilage and other tissues, which reduces credibility of the biomechanical analysis. Innovative application was used by Pang et al. to register several radiologic images and fuse them into a three-dimensional reconstruction of femoral head[14]. By combining X-ray images, CT and MRI image model of individual ONFH patients after images conversion, the radiologic images are registered by Mimic software and then successfully integrated into a model. This multi-images registration technique allows the model to cover cortical bone, cancellous bone, cartilage, necrosis areas and trabecular bone (Figure 2)."

"

Stress analysis in femoral head necrotic area Biomechanical effect of sclerosis rim: Sclerosis band is a common pathologic feature in ONFH around the necrosis area, although its biomechanical effect is seldom to be paid attention[15]. Sclerosis band is a specific mechanical product after ONFH progress with the formation of necrosis, but also the marker of necrosis healing[16]. Natural history of sclerosis formation is less studied. Nevertheless, it can be regarded as a biomechanical support to postpone or prevent ONFH from malformation[17]. Protection of sclerosis band keeps further damage away, including collapse. Yu et al.[18] use finite element analysis to evaluate the stress of sclerosis band at the proximal femur. The study simulated different proportions of sclerosis band in the proximal edge and added loads on it. Series of indicators are ultimately evaluated, like total deformation value, maximum stress, minor stress and the contacted pressure. The results showed sclerosis rim increased gradually with the total value of the femoral head deformation and maximum stress decreasing in turn. Therefore, the authors believe that the proximal edge of the sclerosis band can provide effective mechanical support for the femoral head. Chen et al.[19] hold the same views for sclerosis band. They simulated the stress distribution after bone remodeling and calculated the maximum Von-Mises stress of bone. The results suggest that sclerosis band is an adaptive mechanical product to compensate the elastic modulus of the femoral head. Sclerosis band is an alternative mechanical reinforcing structure, which can significantly reduce the pressure loading in the subchondral bone, but only when the necrosis is small. However, Karasuyama et al.[20] made use of the finite element model to assess stress distribution in fracture zones of ONFH. They found that both necrosis healing and fracture appearing are resulted from the change of sclerosis band.   Stress distribution and collapse Range of necrosis area and stress distribution have a significant impact on the collapse occurring[21]. Stress levels in necrosis area decrease and stress distribution concentrate around the necrotic bone, both result in trabecular fracture and interruption of blood vessel, which will inevitably lead to biomechanical instability in the femoral head[22-23]. Currently, focus of the researches has breakthrough the original base in gross stress analysis, and inclined to study each supporting column in the femoral head. Such kinds of methods in simulating every portion are the innovational tendency to investigate the collapse stress.   Fang et al.[24] established different volumes of necrotic tissue in ONFH by a finite element model, and put on and increase the load on the model (Figure 3). It was found that the range of necrosis area would confirmedly influence stress distribution in femoral head. The stress in the surface is higher than the one in bottom of the necrotic area. The pathologic mechanism can be described that cancellous bone fracture occurs under the subchondral region when the peak stress exceeds a critical value, which is the direct cause of the collapse of the femoral head.   Karasuyama et al.[20] assessed the stress distribution and evaluate the subchondral fracture zone by measuring Von Mises value. In the meantime, the results are combined with the detection of osteoclasts count in necrosis tissue by pathologic methods. The authors finally concluded that the shear stress and shear tension force tend to be concentrated on trabecular site near the thicken border and osteoclast activity is reason to enhance the rate of collapse.   Finite element analysis for therapy Finite element analysis and kinds of hip preservation surgeries Core decompression with multiple drilling, a kind of mature hip preservation surgery, aims to reduce intramedullary pressure, prevent the development of ONFH progress and recover the microcirculation through the hole and tunnel[25-26]. This surgery is relatively simple and easy to operate, even if there are still some technical difficulties[27-28]. For example, a lack of mechanical support, selection of the optimal entry for grafting, suitable number of the drilling holes and so on. Bae et al.[28] studied drilling technique by finite element analysis and investigated the clinical effect of the quantity and entry location of drilling holes. They calculated femoral stress and strain state when the surgeons took the drilling technique. They also analyzed cadavers to assess the best number of the drilling, related trabecular bone density near the entrance according to Von-Mises stress. The results supported that the treatment with multiple drilling is a stable operation. Tran et al. [29] had a similar study in simulating and analyzing core decompression. The results showed that the use of standard drilling core decompression combined with graft implanted will prevent patients from a risk of hip fracture in daily activities. Only when the entry point is at the distal femur, the risk of hip fracture increases. The optimal entry point should be placed at a proximal end of the rotor. Thilo et al.[27] studied core decompression with tantalum rod implant. Moreover, they concluded that core decompression using small drill holes seems to be better compared to the tantalum implant and to conventional core decompression for its low risk of fracture. The results also suggest that a superior performance for core decompression presumably due to the lack of complete bone ingrowth of the tantalum implant. Shi et al.[31] compared the collapse level and collapse ranges under different weights after the patients accepted tantalum rod implant, fibula implant or core decompression. The results showed that smaller necrosis range (60°) is, less different the three operations effect on collapse level. When the necrotic area is in a wide range (120°), all the operations can reduce this level. Lutz et al.[32] compared three kinds of operations by simulating models in finite element analysis, including core decompression, minimal invasive drilling and tantalum rod implant, and then predict their long-term efficacy. Zhou et al.[10] found that the finite element can reconstruct the biomechanical transfer path of the femoral head and reduce the risk of a collapse in the cortical bone (Figure 4).   Hip resurfacing arthroplasty Hip resurfacing arthroplasty is a mature surgical treatment for ONFH[33]. Biomechanical changes surrounding the hip resurfacing arthroplasty, however, have not been completely studied[34]. The studies are focused on individual necrosis range and different insertion angles of the implant. Tai et al.[35] investigated that ONFH patients who accepted hip resurfacing arthroplasty were usually to have fracture near femoral neck in long-term follow-up. This phenomenon is caused by stress shielding. The authors simulated four different degrees of necrotic area (0°, 60°, 100°, 115°) and three different insertion angles of implant (10° vara, neutral hip, 10° valgus) in the finite element model. It was noted that the stress distribution after prosthesis of hip resurfacing arthroplasty occurs in the femoral neck. Sakagoshi et al.[36] evaluated the biomechanical changes of HRA in various degrees of necrosis area and implantation angles. Similarly, the authors established three modes of necrosis range of femoral head surface and designed five hip resurfacing arthroplasty handle angles. The results show that the maximum stress distribution on the implant was decreased with stem angles ranging from 130° to 140°. In addition, the peak stress of cement increased with the stem angle turns to vara. Thus, it is required to be mentioned that a large area of necrosis, valgus or varus prosthesis are likely to worsen the biomechanical environment, even resulting in revision arthroplasty.   Other operations and surgical planning Besides the treatment as described above, there are many other kinds of innovative affiliated technologies used in reducing head stress or supporting collapse in ONFH[37]. Surgeons can also utilize finite element analysis to demonstrate their feasibility. Xiao et al.[11] analyzed a support device for collapse femoral head. According to finite element analysis of the femoral head surface load stress and strain, author reported that stress concentration of normal femoral head model is in the distance of the femur. The model of the implant stress concentration should focus on the lower part of the lesser trochanter of the femur and the displacement of the stress point is smaller than normal. The authors believed that the device can effectively maintain the biomechanical properties of the femoral neck. Zhou et al.[38] use finite element analysis to modulate the process of percutaneous balloon filling with bone cement femoral head necrosis. The steps include simulating the walking by a single leg and hip peak stress; simulating the avascular necrosis model and balloon filling with bone cement model respectively; calculating Von-Mises values at the top of the femoral head, femoral neck and surrounding area. Fresh human avascular necrosis of femoral head specimens for biomechanical analysis showed that percutaneous balloon filling with bone cement can indeed prevent collapse. Pang et al.[39] used finite element analysis to have a quantitative research in hip preservation surgery based on surgical indications and optimized surgical plan the finite element analysis provided. Finite element analysis can calculate the necrotic area, location, size, volume, weight loading area, percentage of necrosis area in three-dimensional visualization and simulate the procedures of core decompression surgery, bone graft supporting surgery, fibula supporting surgery so as to determine and optimize various indexes in the hip preservation program. "