An Editorial for the Special Issue of “Recent Advances in Biomechanics of Human Movement and Its Clinical Applications”

1. Introduction

When we get up in the morning, a question may occur in our brain. What topics or which kinds of research projects should we carry out in our research field? While our labs have various instruments available, e.g., high-speed motion capture systems, force platforms, pressure measurement systems, electromyography and electroencephalogram systems, metabolic measurement systems, etc., this question might be easy to answer. Yes, you may use one of the measurement systems to collect data for a feasible project, and then you will obtain the outcomes quickly. But hold on; another question could occasionally arise in the brain. How do we link our research projects to clinical practice? This question stimulated interest to propose this Special Issue. Volume 1 of the issue comes out and could partially answer the latter question. Obviously, there is not any volume that could completely answer the question. However, this volume showed our efforts to consider our research activities being associated with clinical practices.

In the literature, there have recently been some efforts to dig various aspects of the biomechanical applications related to clinical practice. For example, some studies applied gait analysis to clinical applications, e.g., giving advanced methods for general situations [1,2] and reporting about how the prostheses affected joint loads during gait [3,4]. Some studies mentioned advanced tools and means that could be applied in gait analysis, e.g., artificial intelligence, machine learning, and wearable sensors [5,6]. Regarding finite element analysis, some studies have further applied the tool into surgery and clinical practice [7,8,9]. Though many methods have been proposed by some reviewing articles [5,10], there is a lack of effective applications of those intentions into clinical practice. In other words, there are many big gaps between potential research activities and clinical applications.

It is pleasing to see that Volume 1 of the Special Issue was published, although this volume showed only a corner of the wide research field.

2. An Overview of Published Articles

This volume has collected 10 articles on different aspects of the current research field. There are four articles related to gait analysis, some for patient cases and some for healthy groups; two articles on instrumentation, either for marker-less motion capture systems or for measuring eye-related tension forces; one paper on finite element modeling for surgical skill; one work on theoretical modeling and validation in stability analysis; one study on the relationship between humans and robots; and one article on a general description of kinematics and kinetics for squatting. These articles have reflected current progress in the research field, and thus it is worth being briefed as below.

Harro et al. (Contributor 1) investigated whether there would be differences in dynamic postural control during self-paced walking and obstacle-crossing between the people with Parkinson’s disease (PD) and the healthy control group. They found a significantly reduced step length for the trailing limb and a significantly larger mediolateral distance between the center of pressure and center of mass during first double support and second double support for the second step. Their results suggested that the margin of stability may be useful in discerning postural control adaptations during obstacle negotiation in the PD group.

Similarly, Alderink et al. (Contributor 2) studied the risk of falls for PD, especially in turning. Their project investigated the dynamic balance control in people with PD compared to the healthy individuals during walking with 90° turns. The results showed that there were no significant differences in the gait variables or dynamic balance parameters between the two groups. This finding indicated that the people with PD in the early stages of disease may not be seen as different from the healthy group. It should be noted that a future study on the people with PD in later stages should be considered.

Using the entropy concept, Tipton et al. (Contributor 3) investigated the body’s postural control system with balance characteristics. They collected the center of pressure for stances with feet together and feet tandem, with eyes opened or closed, in neurotypical participants. The parameters included the indices of entropy and velocity to examine how sensitive the variables were tracking changes at stability levels. The results showed increased approximate entropy in anterior/posterior and medial/lateral directions and increased velocity in both directions during the conditions of decreased stability. This study indicated that researchers could use motion analysis to combine with the concepts from other fields, e.g., signal analysis, to obtain useful information for clinical applications. Given that motion analysis is mainly on physics and biomedical engineering, the other fields may include physiology, artificial intelligence (AI), robotics, anatomy, etc.

The paper by Alsirhani et al. (Contributor 4) is on both gait analysis and posture maintenance. They checked if a long time of cross-leg sitting (CLS) affected gait. The results showed that some gait parameters, e.g., walking speed, are improved while kinetic parameters, e.g., joint forces, are not changed too much by CLS. As the CLS is a habit posture in some Asian countries, this study indicated a strongly social significance.

Ahn et al. (Contributor 5) examined a marker-less motion capture system and compared it with the Vicon^® motion capture system, the latter being considered as one of the best systems in motion analysis. They collected basic movements, e.g., hip flexion, knee flexion, and trunk rotation using both the new and Vicon systems simultaneously, from five healthy participants. To assess the systems, the intra-class correlation coefficient (ICC) was calculated to examine angular-range validity and the intra-joint reliability of the systems. The results showed that the ICCs for hip flexion and knee flexion were 0.924–0.998 and 0.842–0.989, respectively. They concluded that the new markerless motion capture system had a reasonable reliability in measuring joint kinematics during movements. However, the results also showed significant differences between the new and Vicon systems in some parameters and curves. From their paper, though it is encouraging to see the development of markerless systems, it is recognized that there is a long way from the development of markerless systems in the labs to the application in clinical practice.

Shin et al. (Contributor 6) developed a tiny instrument to measure the tension in the extraocular muscles (EOM), which functionally control eyeball movements. The abnormal tension of EOM may cause strabismus. The device is non-invasive, able to evaluate the active EOM tension. Their design used a simple cotton-tipped swab to pass the EOM tension to a force sensor. The changing angle of the swab and the force of EOM tension were wirelessly passed to the software, and then the measured values were displayed in real-time. The measured force ranges covered from several to hundreds of grams. The proposed instrument is simple, practical, and lower cost compared to the traditional ones. The device is not only for applications of strabismus measurements but also provides a good example in clinical instrumentation.

Liu et al. (Contributor 7) investigated the reachable workspaces by the upper limbs under different visual and physical conditions and compared the movement efficiencies with and without the help of robots. Ten subjects participated in an experiment, where there were various reaching tasks required. The durations to complete the tasks were recorded for the two groups, one human only (H) and another human–robot-collaborated (HRC). From the data collected, this study statistically divided the workspace into five ranges (I to V), with the angles being 0°~44°, 44°~67°, 67°~81°, 81°~153°, and 153°~180° from the anterior sagittal plane. The results showed that H and HRC have similar outcomes in the range I-II, but the HRC group was significantly faster than the H group in the range III to V, by reduced times of roughly 87%, 70%, and 60%. This study provided evidence such as (1) reachable workspace ranges and (2) in which ranges HRC performed better. The limitation was that the tasks did not include complex handling jobs that humans can complete easily but the robot cannot. In other words, more advanced robots should be developed in the future.

While most studies use a three-dimensional (3D) model currently, Sopa et al. (Contributor 8) proposed a two-dimensional (2D) model for analyzing human stability. The model was able to estimate the force and moment at the joints using a closed kinematic chain and the Euler–Newton approach. The model still required the data from a motion capture system, a force plate, and a dynamometer. This study also included experiments involving two activities for maintaining a state of equilibrium. Their results showed the differences between the two movements. This study indicated that researchers are encouraged to apply some simplified models to analyze motion in clinical practice.

As a report, Hoogenboom et al. (Contributor 9) described the 3D kinematics and kinetics for healthy participants during an overhead deep squat with 70 healthy adults aged 18–35 years old. They employed a Vicon motion capture system with force plates. The full-body 33 markers from the Plug-in-Gait model were used. The reported parameters included the lower limb and trunk joint angles in the sagittal, front, transverse planes, the ground reaction forces and joint moments, etc. The results showed that the highest joint moments were in the sagittal plane, while other planes maintain stability. These results can be used as a reference for the clinicians who assess biomechanical abnormalities for patients and prescribe reasonable exercises in rehabilitation settings.

Andrade et al. (Contributor 10) proposed a 3D finite element model to investigate which blade position was optimum for the cut-off in fracture fixation implants. Given the fact that there are many difficulties that surgeons face in clinical surgery, this topic is helpful. The study proposed a stiffness-adaptive damage model that could evaluate the risk of cut-out. Gait loading was used in the assessment of the eight positions in terms of the relative risk of cut-out for each. The results suggested that the closer the tip of the blade to the femoral head surface, the lower the risk of cut-out. The results also showed that the blade in the medial–lateral direction and its superior–inferior position had influence on the risk of cut-out. This study is so helpful that surgeons may consider their treatment plan and improve their skill accordingly, although optimal blade positioning is subject-specific, depending on various clinical factors.

3. Conclusions

According to the number of articles published in each area, 40% of the papers are on gait analysis, 20% on instrumentation, 10% on computer modeling, 10% on the relationship between human and human–robot collaboration, 10% on general kinematics and kinetics, and 10% on finite element analysis. Basically, these proportions may have partially reflected various aspects of the current research field. However, these proportions do not reflect the tendency in future research. Potentially, there are still many areas for researchers to explore in the future. In the introduction of this Special Issue, there was a list of key words and topics indicating research directions, but there were not any articles in some of the proposed research topics. For example, there is no paper on artificial intelligence, machine learning, and deep learning; motion pattern recognition; bio-signal measurement and analysis; or bio-image measurement and analysis collected in this volume. One reason could be that this volume was open for a short period of time and thus researchers in these fields may not have recognized it while other special journals may have attracted potential authors to publish their articles there. Another reason may be that the research activities in these proposed topics are less active than in other fields, such as those of the articles published in Volume 1.

In terms of key words, there were no studies on electromyographs (EMGs) and electroencephalograms (EEGs) in this volume. In clinical practice, these electronic signals are so important as they are only a way to detect muscle and brain activities. For the patients with cerebral palsy and stroke, the analysis and interpretation of EMG and EEG may help us understand their movements deeply and then provide potential clues in clinical treatments and rehabilitation settings. It is time for researchers to pay more attention to projects on EMG and EEG.

Another big and popular research direction is on artificial intelligence (AI). Compared to the progress in other subjects, AI application in biomechanics associated with clinical practice is relatively lacking. A major reason could be the difficulty in collecting data. In general, AI and machine learning require a significant amount of data as the base for modeling. It would be impossible to make reasonable AI models without the support of a large sample size. It is recognized that collecting data in a national or worldwide health system, i.e., hospitals and clinics, requires strict ethical permissions. On the other hand, researchers doing AI modeling may not clearly know which data they need and do not need at the beginning. Therefore, an ethics application may not be able to provide a detailed data collection plan. This may prevent researchers from obtaining useful data.

Following the two aspects above, a key problem to be solved could be algorithms. Once you have both data and general AI methods, a suitable algorithm is needed to combine two things together. The design and testing of algorithms require not only a lot of time but also creative abilities. A successful algorithm could not only produce interesting results from the collected data but also link the research outcomes to clinical applications quickly. Usually, clinicians cannot make their own application directly from the data they collected. Therefore, it is necessary for researchers in biomedical engineering, mathematics, computing, biology, or similar fields to make a bridge between useful data and clinical applications. A good algorithm could bring in not only a good interpretation of clinical data but also a good application. Therefore, a creative algorithm could largely drive the research activities forward as shown in Figure 1.

As a whole, this editorial is missing many potential research directions while highlighting a few of them. It is certain that novel ideas and projects will be proposed by the clinicians and academic researchers who work together. Last, but not least, this Special Issue is considered the stage where papers exploring new research directions and benefiting clinical practice by using biomechanical principles and updated tools were warmly welcome. The aim of this Special Issue has been and is to find interesting studies in human movement and its clinical applications.