Point cloud completion is a generation and estimation issue derived from the partial point clouds, which plays a vital role in the applications of 3D computer vision. The progress of deep learning (DL) has impressively improved the capability and robustness of point cloud completion. However, the quality of completed point clouds is still needed to be further enhanced to meet the practical utilization. Therefore, this work aims to conduct a comprehensive survey on various methods, including point-based, view-based, convolution-based, graph-based, generative model-based, transformer-based approaches, etc. And this survey summarizes the comparisons among these methods to provoke further research insights. Besides, this review sums up the commonly used datasets and illustrates the applications of point cloud completion. Eventually, we also discussed possible research trends in this promptly expanding field.

Hyperspectral image-anomaly detection (HSI-AD) has become one of the research hotspots in the field of remote sensing. Because HSI’s features of integrating image and spectrum provide a considerable data basis for abnormal object detection, HSI-AD has a huge application potential in HSI analysis. It is difficult to effectively extract a large number of nonlinear features contained in HSI data using traditional machine learning methods, and deep learning has incomparable advantages in the extraction of nonlinear features. Therefore, deep learning has been widely used in HSI-AD and has shown excellent performance. This review systematically summarizes the related reference of HSI-AD based on deep learning and classifies the corresponding methods into performance comparisons. Specifically, we first introduce the characteristics of HSI-AD and the challenges faced by traditional methods and introduce the advantages of deep learning in dealing with these problems. Then, we systematically review and classify the corresponding methods of HSI-AD. Finally, the performance of the HSI-AD method based on deep learning is compared on several mainstream data sets, and the existing challenges are summarized. The main purpose of this article is to give a more comprehensive overview of the HSI-AD method to provide a reference for future research work.

It has been hypothesized that diabetic foot ulceration might be internally initiated. This study established a three-dimensional and nonlinear finite element model of a human foot complex with comprehensive skeletal and soft-tissue components. The model was validated by experimental data of subject-specific barefoot plantar pressure measurements. The feasibility of the model to predict the 3D, internal, plantar soft-tissue deformation and stress was evaluated. The preliminary results indicate that large von Mises stress occurs where plantar soft-tissue contacts with geometrically irregular bony structures, thus internal stress distribution within the plantar soft-tissue was dramatically influenced by bony prominences due to stress concentration. At the forefoot part, an average stress magnification factor of 3.01 was quantified. The lateral sesamoid bone associated to the 1st MTH showed the largest effect of stress concentration. The modeling approach presented provides a possible way to explore the complexity of the mechanical environment inside the plantar soft-tissue.

Summary: Musculoskeletal injuries are a common cause of lost training days and wastage in racehorses. Many bone injuries are a consequence of repeated high loading during fast work, resulting in chronic damage accumulation and material fatigue of bone. The highest joint loads occur in the fetlock, which is also the most common site of subchondral bone injury in racehorses. Microcracks in the subchondral bone at sites where intra‐articular fractures and palmar osteochondral disease occur are similar to the fatigue damage detected experimentally after repeated loading of bone. Fatigue is a process that has undergone much study in material science in order to avoid catastrophic failure of engineering structures. The term ‘fatigue life’ refers to the numbers of cycles of loading that can be sustained before failure occurs. Fatigue life decreases exponentially with increasing load. This is important in horses as loads within the limb increase with increasing speed. Bone adapts to increased loading by modelling to maintain the strains within the bone at a safe level. Bone also repairs fatigued matrix through remodelling. Fatigue injuries develop when microdamage accumulates faster than remodelling can repair. Remodelling of the equine metacarpus is reduced during race training and accelerated during rest periods. The first phase of remodelling is bone resorption, which weakens the bone through increased porosity. A bone that is porous following a rest period may fail earlier than a fully adapted bone. Maximising bone adaptation is an important part of training young racehorses. However, even well‐adapted bones accumulate microdamage and require ongoing remodelling. If remodelling inhibition at the extremes of training is unavoidable then the duration of exposure to high‐speed work needs to be limited and appropriate rest periods instituted. Further research is warranted to elucidate the effect of fast‐speed work and rest on bone damage accumulation and repair.

Recently, numerous types of posterior dynamic stabilization (PDS) devices have been introduced as an alternative to the fusion devices for the surgical treatment of degenerative lumbar spine. It is hypothesized that the use of 'compliant' materials such as Nitinol (Ni-Ti alloy, elastic modulus = 75 GPa) or polyether-etherketone (PEEK, elastic modulus = 3.2 GPa) in PDS can restore stability of the lumbar spine without adverse stress-shielding effects that have often been found with 'rigid' fusion devices made of 'rigid' Ti alloys (elastic modulus = 114 GPa). Previous studies have shown that suitably designed PDS devices made of more compliant material may be able to help retain kinematic behavior of the normal spine with optimal load sharing between the anterior and posterior spinal elements. However, only a few studies on their biomechanical efficacies are available. In this study, we conducted a finite-element (FE) study to investigate changes in load-sharing characteristics of PDS devices. The implanted models were constructed after modifying the previously validated intact model of L3-4 spine. Posterior lumbar fusion with three different types of pedicle screw systems was simulated: a conventional rigid fixation system (Ti6Al4V, Phi = 6.0 mm) and two kinds of PDS devices (one with Nitinol rod with a three-coiled turn manner, Phi = 4.0 mm; the other with PEEK rod with a uniform cylindrical shape, Phi = 6.0 mm). To simulate the load on the lumbar spine in a neutral posture, an axial compressive load (400 N) was applied. Subsequently, the changes in load-sharing characteristics and stresses were investigated. When the compressive load was applied on the implanted models (Nitinol rod, PEEK rod, Ti-alloy rod), the predicted axial compressive loads transmitted through the devices were 141.8 N, 109.8 N and 266.8 N, respectively. Axial forces across the PDS devices (Nitinol rod, PEEK rod) and rigid system (Ti-alloy rod) with facet joints were predicted to take over 41%, 33% and 71% of the applied compression load, respectively. Our results confirmed the hypothesis on the PDS devices by showing the substantial reduction in stress-shielding characteristics. Higher axial load was noted across the anterior structure with the PDS devices, which could slow the degeneration process of bony structures and lower the possibility of implant failure.