Motion capture (mocap) systems are extensively utilized in healthcare for monitoring rehabilitation programs, facilitating clinical gait assessments for early Alzheimer’s diagnosis, managing walking disorders, and developing exoskeleton suits. However, like many other healthcare technologies, mocap systems have some flaws, like missing markers and occlusions. Given mocap data’s sequential and temporal nature, understanding marker relationships and capturing global dependencies are crucial for effective human motion recovery applications. To address these challenges, we proposed an unsupervised transformers framework for human motion recovery, called IMU-Trans. We evaluated our framework’s generalizability across two clinical datasets and tested its robustness by adjusting the missing marker rates, comparing its performance against low-dimensional Kalman filtering, long short-term memory (LSTM), and gated recurrent unit (GRU) models. Our experimental results demonstrated that IMU-Trans outperforms state-of-the-art models by training in an unsupervised manner. The closest competitor, GRU, demonstrated an RMSE of 1.35 ± 0.82, 2.36 ± 1.26, 3.43 ± 1.73, and 4.39 ± 2.18 cm for 20

This study develops an automatic defect detection system for polyvinyl chloride (PVC) profile manufacturing, addressing inefficiencies in manual inspection. It compares the proposed autoencoder model with other well-known unsupervised deep-learning methods, including GANomaly, f-AnoGAN, and the student-teacher network, for defect detection during extrusion. Utilising a defective PVC profile dataset, the study generates anomaly heat maps through reconstruction errors and assesses model performance using the area under the receiver operating characteristic (ROC) curve. The proposed autoencoder model is found to be optimal for this dataset, offering a balance between efficiency and accuracy. These findings have significant implications for enhancing quality control and reducing defects in PVC manufacturing, with potential applicability in other industrial settings.

This study presents an alternative method of normalizing Surface Electromyography (EMG) signals, which provide valuable insights into the musculoskeletal properties of the human body. Traditional normalization through the peak maximum voluntary contraction (MVC) method may result in variable outcomes due to its dependence on pre-processing steps such as smoothing and window length. To address this issue, the study standardizes the EMG pre-processing steps by using non-linear techniques, namely recurrence quantification analysis (RQA) and heterogeneous RQA (HRQA). RQA, specifically HRQA, functions as EMG feature extractors to explore the nonlinear dynamical system and state space trajectories of chaotic EMG signals. Principal component analysis (PCA) and kernel PCA (KPCA) were further applied to the RQA and HRQA results to obtain data with more information. The results show that the first principal components (PC1 and KPC1) of HRQA have a very strong relationship with peak MVC ( rho > |0.90|, p < 0.001), indicating their potential compared to the traditional EMG normalization technique for lower limb muscle groups. Additionally, the study investigated the dynamic differences among isometric MVC movements in lower limb muscle groups and found that the RQA and HRQA methods indicated a greater separation for detecting these differences compared to peak MVC. This study offers a valuable method for assessing the musculoskeletal characteristics of individuals who may have difficulty with activities like walking or providing maximum force during EMG data collection, such as those suffering from Parkinson ' s disease or walking disabilities. This enhances our understanding of their musculoskeletal properties.

In U.S. higher education institutions, there's an increasing focus on attracting more students, especially minorities and underrepresented groups, to engineering majors. Despite existing research on factors influencing engineering major choices, the specific roles of demographics, such as gender and race, remain crucial for enhancing diversity and inclusivity in engineering departments. This study investigates how and when undergraduate engineering students prioritize various factors in selecting their majors, with a particular focus on differences across gender, race, and specific majors. Utilizing Ecological Systems Theory (EST) as a framework, the study analyzes responses from 276 engineering students at a public university in the Midwest during the 2022/2023 academic year, employing methods like crosstabs, chi-square, t-tests, ANOVA, and regression analyses. Results indicate that students' priorities align with EST systems, considering aspects like salary, influence from family and friends, and guidance from advisors and professors, as well as the importance of hands-on projects. Notable variations are observed among demographic groups: White, Female/Genderqueer (FGQ) students, and those in Industrial Engineering (IE), Construction Management (CM), and Mechatronics and Robotics Engineering (MRE) majors emphasize salary, while American Asian or Pacific Islander (AAPI) students value hands-on projects more. Additionally, FGQ students find advisors more influential, whereas AAPI students lean more towards professors. These findings highlight the necessity for tailored support mechanisms in engineering schools, particularly addressing the unique challenges and needs of FGQ, AAPI, and Students of Color (SOC) students, and emphasize the importance of personalized guidance to facilitate informed major selection.

This study investigates the perceived familiarity with different engineering disciplines among engineering students atSouthern Illinois University, Edwardsville (SIUE), with a focus on disparities across gender, race, and academic majors.Data were collected via an online Qualtrics survey from undergraduate engineering students at SIUE during the Fall 2022and Spring 2023 semesters. The survey, which achieved a 25% response rate with 275 completed surveys, assessed students'ability to identify six key engineering disciplines. Quantitative analysis methods, including linear regression andmultivariate analysis, were applied to examine how demographic factors influence students' familiarity and identificationskills. Weighted data were used in the analysis to correct for non-response bias related to gender, race, and major selection.Findings indicate notable disparities: male and White students exhibited higher familiarity compared to Female andGender Queer (FGQ) students and students of color, especially Asian American and Pacific Islander (AAPI) students.Students enrolled in broad-based and interdisciplinary engineering programs demonstrated better overall understandingthan their peers in more specialized programs. The study confirms that perceived familiarity significantly predicts correctidentification, suggesting that increased educational exposure to various engineering fields can enhance accuracy. Theseresults underscore the importance of revising educational curricula to include more inclusive and comprehensive exposureto all engineering disciplines. Interventions such as redesigned coursework, expanded mentorship programs, andincreased support for underrepresented groups are recommended to address the observed disparities. Future researchshould aim to validate these interventions and explore longitudinal trends in engineering education familiarity

This paper examines the potential benefits and risks of using ChatGPT, an AI-powered chatbot developed by OpenAI, in engineering education by generating sample questions and answers and solving sample mathematical problems related to course subjects in the industrial engineering curriculum.The generated questions and answers were evaluated using multiple criteria such as accuracy, relevance, precision, depth, and breadth.The examination found that ChatGPT has the potential to be a useful tool for generating questions and answers in various contexts.However, the study also raises concerns about the potential risks of using ChatGPT in engineering education, such as over-reliance on the chatbot leading to a decline in critical thinking and problem-solving skills, and the potential for academic dishonesty and misinformation, as well as confusion among students due to the generation of wrong answers to mathematical problems which can negatively impact students' learning experience.The author emphasizes the importance of educators being aware of these potential risks and developing strategies and policies to ensure ChatGPT is used as a supplement, not a replacement, for traditional teaching methods.

This paper examines the potential uses of ChatGPT in generating assessment tasks that can be used across different disciplines in higher education. To illustrate this, we provide examples from three courses in the disciplines of industrial engineering and applied linguistics: Project Analysis and Control, Manufacturing Processes, and Introduction to Linguistics. Our examination of ChatGPT focuses on how some of the common, but time-consuming, tasks can be generated by using ChatGPT as a supportive instructional tool. We observed in our analysis that ChatGPT demonstrates a high level of performance in generating assessment questions and tasks that are accurate and on-topic, and a level of creativity and flexibility in its question generation capabilities. However, it is important to note that ChatGPT is not designed to replace human expertise or judgment. It is crucial that instructors carefully evaluate the reliability and accuracy of the assessments or information generated by ChatGPT.

Recent advancements in quality control across various industries have increasingly utilized the integration of video cameras and image processing for effective defect detection. A critical barrier to progress is the scarcity of comprehensive datasets featuring annotated defects, which are essential for developing and refining automated defect detection models. This systematic review, spanning from 2015 to 2023, identifies 15 publicly available datasets and critically examines them to assess their effectiveness and applicability for benchmarking and model development. Our findings reveal a diverse landscape of datasets, such as NEU-CLS, NEU-DET, DAGM, KolektorSDD, PCB Defect Dataset, and the Hollow Cylindrical Defect Detection Dataset, each with unique strengths and limitations in terms of image quality, defect type representation, and real-world applicability. The goal of this systematic review is to consolidate these datasets in a single location, providing researchers who seek such publicly available resources with a comprehensive reference.

A comprehensive investigation of various maximum voluntary contraction (MVC) positions to determine the optimal positions for vastus lateralis (VL), biceps femoris (BF), gastrocnemius lateralis (GL), and tibialis anterior (TA). Twelve participants performed total of seventeen MVC positions for major lower limb muscle groups (VL, BF, GL, and TA). Neuromuscular activities were recorded by surface electromyography. Signals were smoothed by root mean square (RMS). Each MVC level were expressed as a percentage of MVC (% MVC). Statistical differences were measured with a one-way repeated measures analysis of variance and a Tukey’s HSD (p < 0.05). Optimal MVC positions were found as follows: (i) VL: the combination of knee extension at 70° and 90° flexed knee in sitting position; (ii) BF: the combination of knee flexion at 30°, 45°, and 60° flexed knee in prone position; (iii) GL: unipedal standing position; (iv) TA: the combination of dorsiflexion in sitting position and ankle neutral, in standing position and ankle 110°, and in standing position and ankle 70°. This study confirms that multiple positions were needed to elicit the maximal MVC values for VL, BF, and TA. For GL, single MVC position should be performed to elicit the maximal MVC.

BACKGROUND:Over the last decade, there has been a steady increase in the number of children diagnosed with autism spectrum disorder (ASD) on a global scale, impacting all racial and cultural groups. This increase in the diagnostic rate has prompted investigation into a myriad of factors that may serve as early signs of ASD. One of these factors includes the biomechanics of gait, or the manner of walking. Although ASD is a spectrum, many autistic children experience differences in gross motor function, including gait. It has been documented that gait is also impacted by racial and cultural background. Given that ASD is equally prevalent across all cultural backgrounds, it is urgent that studies assessing gait in autistic children consider the impact of cultural factors on children's development of gait. The purpose of the present scoping review was to assess whether recent empirical research studies focusing on gait in autistic children have taken culture into account. METHODS:To do so, we conducted a scoping review following PRISMA guidelines using a keyword searching with the terms autism, OR autism spectrum disorder, OR ASD, OR autis, AND gait OR walking in the following databases: CINAHL, ERIC (EBSCO), Medline, ProQuest Nursing & Allied Health Source, PsychInfo, PubMed, and Scopus. Articles were considered for review if they met all six of the following inclusionary criteria: (1) included participants with a diagnosis of autism spectrum disorder (ASD), (2) directly measured gait or walking, (3) the article was a primary study, (4) the article was written in English, (5) participants included children up to age 18, and (6) the article was published between 2014 and 2022. RESULTS:A total of 43 articles met eligibility criteria but none of the articles took culture into account in the data analysis process. CONCLUSIONS:There is an urgent need for neuroscience research to consider cultural factors when assessing gait characteristics of autistic children. This would allow for more culturally responsive and equitable assessment and intervention planning for all autistic children.

Clinical gait analysis is a useful tool for assessing a patient's walking conditions. Force platforms are gait analysis tools used to collect the ground reaction forces (GRFs); however, they are expensive and time-consuming. Therefore, this study focuses on the prediction of GRFs and joint moments without using force platforms. To address this problem, we proposed to combine deep learning methods with regression transfer learning (RTL). The inputs of the proposed method are joint angles and marker trajectories from a public dataset. Principal component analysis (PCA) has been used to reduce the data dimensionality to improve the computational time and prediction accuracy. A synthetic dataset has been generated to pre-train the deep learning method for transfer learning purpose. The experimental results indicate that the proposed transfer learning method increases the target domain's learning process and can successfully predict the average GRFs and joint moments with 97.44% and 96.56% accuracy, respectively.

Gait analysis compares the gait characteristics of people with health issues to those of a control group in order to detect gait abnormalities. This comparison is carried out by evaluating a number of gait parameters with discrete values. Gait data, on the other hand, is time-series data and must be assessed using a different approach. The purpose of this study was to develop a quantitative measure that takes into account time-series data for comparing the gait characteristics of two groups of individuals using clustering. The gait data were collected using an optical motion capture system. An adaptive density-peaks clustering technique with a shape-based similarity measure was employed to compare gait characteristics. The results demonstrate that the proposed adaptive density-peaks clustering technique, which employs dynamic derivative time wrapping distance measurement, outperforms three state-of-the-art clustering algorithms for comparing the gait characteristics using time-series gait data.

PURPOSE: The aim of this study was to compare different biomechanical models to identify the take-off velocity (TOV) of a vertical jump. METHODS:Fourteen young adults (age = 24 ± 4 yrs) participated in this study. Participants did five maximal vertical jumps while swinging their arms during the jump. Kinematic data was recorded using a 17-marker whole body set and a 10 camera motion capture system. Kinetic data was recorded using a force platform. The data was then analyzed using two kinetic and two kinematic models for phase identification. Both kinetic models (K1 and K2) required the double integration of the ground reaction force to estimate the vertical movement of the center of mass (COM). One kinematic model used 14 markers to create a segmental model (S1) and the other model used 3 markers to create a sacral model (S2) to estimate the movement of the COM. All models defined the start of the eccentric phase as the point where the COM starts to move downward. Three of the models (K1, S1, & S2) defined end of the eccentric phase/start of concentric phase as when the COM starts to move upward. The 4th model (K2) defined the end of the eccentric phase when the velocity of the COM is zero and the start of the eccentric phase when velocity of COM becomes positive. The concentric phase ended when both feet had left the ground for all models. TOV was defined as the velocity of the COM at the end of the eccentric phase. A 1-way ANOVA was used to identify differences in TOV between the 4 different models. RESULTS: There were significant differences between TOV for the 4 models (p < 0.001). Post-hoc comparisons show that the estimated TOV for S1 (3.21 ± 0.43 m/s) was significantly greater than K1, K2, and S2 (2.79 ± 0.43, 2.79 ± 0.43, and 2.67 ± 0.52 m/s respectively). The estimated TOV for S2 (2.67 ± 0.52 m/s) was significantly less than K1 and K2 (2.79 ± 0.43 and 2.79 ± 0.43 respectively). CONCLUSIONS: These results indicate that a whole body kinematic model (S1) results in a greater estimated TOV when compared to kinematic model that doesn’t account for whole body movement (S2). A whole body kinematic model also results in greater estimated TOV when compared to kinetic models.

Flipped teaching (FT) has gained attention due to its method of teaching that allows students to learn basic concepts on their own using instructor‐provided resources before their scheduled class time. The class time is used in rehearsing and applying the knowledge using active learning approaches. The COVID‐19 pandemic caused an unexpected shift from face‐to‐face to remote learning during the middle of the spring 2020 semester challenging both students and faculty. This study examined the transition of students from the flipped classroom method to rapid online learning and whether the transition was dependent on the faculty experience with FT. The perception of the transition of students in the classrooms of faculty (n=12) with extensive experience in FT (cohort 1) was compared with that of a second cohort (n=11) who were in their first semester of FT implementation. Both qualitative and quantitative survey data were collected from 23 classrooms (n=256 students). It was found that the students in the courses taught by cohort 1 who had received intensive FT training and implemented FT in semesters prior to the COVID‐19 pandemic were able to facilitate a smoother transition and adjustment to fully online learning for their students compared to the students in the courses taught by cohort 2 (p< 0.01). The qualitative data analysis suggested that the students participating in the FT courses in the first half of the semester, before the pandemic, had an easier transition to fully online learning. Students also expressed concerns that online learning was affected by the lack of interaction with faculty and peers, lack of motivation, issues with time management, and personal and technological demands. In conclusion, this study found that faculty experience with FT helped ease the transition of students from the face‐to‐face FT to the online format during the COVID‐19 pandemic.

Scheduling problems have been studied extensively over the years. Broad ranges of focus areas exist, from optimization to heuristics, production planning to production sequencing, customized algorithms to general-purpose algorithms, and from simple machines to complex environments. In this research, a heuristic approach has been proposed to overcome the scheduling problem on a complex job shop found at a manufacturer of commercial building products. The research is aimed at sequencing production orders in near-real-time, primarily to minimize total tardiness, but also to reduce total setup time. A layered Shifting Bottleneck Procedure is employed, with the top layer determining release dates and due dates for individual jobs, and the bottom layer applying algorithms to individual work centers. The outcome of this research is a better production schedule than current methods with minimal computation cost. The proposed framework performs well and could be applied to other production areas.

OBJECTIVES:Cone-beam computed tomography (CBCT) scans enable quantification of interproximal bone loss after implant procedures in dental patients. In order for this quantification to be accurate, software is typically used to manipulate image sets captured before and after implantation to obtain their exact registration (i.e., alignment). However, no affordable CBCT image registration software is currently available for dental applications. Thus, the aim of the present study was to develop a freely available graphical user interface, called DentIR, that automates 2-dimensional (2-D) or 3-D image registration for use in planning dental treatment.METHODS:The DentIR app was designed using the MATLAB environment, downloaded to a desktop personal computer (PC and Mac), and tested for its ease of use and alignment accuracy in the absence of the MATLAB environment.RESULTS:The DentIR app enabled previewing of the CBCT images in 3-D to allow for filtering of each frame to reduce noise and blurring before registration. The 2-D or 3-D registration was tested with four transformation methods. The accuracy of each method was assessed by comparing the mean squared error and the peak signal-to-noise ratio values that were provided by the DentIR app. The registered images could be saved as Portable Network Graphics (PNG) images.CONCLUSIONS:The free, user-friendly DentIR app was easily downloadable to Mac or PC platforms. It provided accurate image registration to aid in the planning of dental treatment. Future updates of the DentIR app include adding the ability to register more than two images at once, enhancing image editing options and enabling registration of a cropped portion of the image for more in-depth analyses.

Although runners are at high risk of back and lower extremity injuries, available tools detect only current injury. Here, a model was developed to analyze kinetic and kinematic running gait data collected by an optical motion capture system to predict future injuries based on an individual's running gait pattern. The two key points, when the joints are most vulnerable because internal forces are the greatest, in the continuous running gait cycle were used to extract average parameter values to create predictive models: the heel strike, and when one leg supports the body weight. Three different prediction models—logistic regression, random forest, and boosting—were built using 10 significant parameters identified in a two-step feature selection approach. All collected metric data were normalized before building the predictive models to avoid outlier values and redundancy. The three models were tested to determine whether they could predict that a participant would incur chronic running injuries in the future based on their current running gait pattern. The logistic regression model had the highest prediction accuracy: the area under the curve was 0.9016 [95% confidence interval (CI) 0.8808–0.9369] for logistic regression, 0.8892 (95% CI 0.8463–0.9152) for the random forest, and 0.8732 (95% CI 0.8401–0.9178) for boosting. Further model development may not only enable clinicians to integrate injury intervention into running programs but also lead to predictive models that recognize patterns associated with neurological disorders, such as Parkinson's disease, autism, and multiple sclerosis, in which gait and balance deficiencies may be symptoms or even predictors of disease.

Precise object boundary detection for automatic image segmentation is critical for image analysis, including that used in computer-aided diagnosis. However, such detection traditionally uses active contour or snake models requiring accurate initialization and parameter optimization. Identifying optimal parameter values requires time-consuming multiple runs and provides results that vary by user expertise, limiting the use of these models in high-throughput or real-time situations. Thus, we developed a nonparametric snake model using an interior point search method applied in iterations to find and improve the set of snake points forming the edge of a shape. At each iteration, one or more snake points are replaced by others in the edge map. We validated the model using binary and continuous edge images of single and multiple objects, and noisy and real images, comparing the results to those obtained using traditional snake models. The proposed model not only provides better results on all image types tested but is more robust than traditional snake models. Unlike traditional snake models, the proposed model requires no user interaction for initializing snakes and no preprocessing of noisy images. Thus, our method offers robust automatic image segmentation that is simpler to use and less time-consuming than traditional snake models.

The automatic extraction of the vertebra's shape from dynamic magnetic resonance imaging (MRI) could improve understanding of clinical conditions and their diagnosis. It is hypothesized that the shape of the sacral curve is related to the development of some gynecological conditions such as pelvic organ prolapse (POP). POP is a critical health condition for women and consists of pelvic organs dropping from their normal position. Dynamic MRI is used for assessing POP and to complement clinical examination. Studies have shown some evidence on the association between the shape of the sacral curve and the development of POP. However, the sacral curve is currently extracted manually limiting studies to small datasets and inconclusive evidence. A method composed of an adaptive shortest path algorithm that enhances edge detection and linking, and an improved curve fitting procedure is proposed to automate the identification and segmentation of the sacral curve on MRI. The proposed method uses predetermined pixels surrounding the sacral curve that are found through edge detection to decrease computation time compared to other model-based segmentation algorithms. Moreover, the proposed method is fully automatic and does not require user input or training. Experimental results show that the proposed method can accurately identify sacral curves for nearly 91% of dynamic MRI cases tested in this study. The proposed model is robust and can be used to effectively identify bone structures on MRI.