Evaluation of Convolutional Neural Network-Based Posture Identification Model of Older Adults: From Silhouette of Sagittal Photographs

Markova V., Markov M., Petrova Z., Filkova S.

Prolonged static sitting at the workplace is considered one of the main risks for the development of musculoskeletal disorders (MSDs) and adverse health effects. Factors such as poor posture and extended sitting are perceived to be a reason for conditions such as lumbar discomfort and lower back pain (LBP), even though the scientific explanation of this relationship is still unclear and raises disputes in the scientific community. The current study focused on evaluating the relationship between LBP and prolonged sitting in poor posture using photogrammetric images, postural angle calculation, machine learning models, and questionnaire-based self-reports regarding the occurrence of LBP and similar symptoms among the participants. Machine learning models trained with this data are employed to recognize poor body postures. Two scenarios have been elaborated for modeling purposes: scenario 1, based on natural body posture tagged as correct and incorrect, and scenario 2, based on incorrect body postures, corrected additionally by the rehabilitator. The achieved accuracies of respectively 75.3% and 85% for both scenarios reveal the potential for future research in enhancing awareness and actively managing posture-related issues that elevate the likelihood of developing lower back pain symptoms.

Sugiyama N., Kai Y., Koda H., Morihara T., Kida N.

Postural assessment is one of the indicators of health status in older adults. Since the number of older adults is on the rise, it is essential to assess simpler methods and automated ones in the future. Therefore, we focused on a visual method (imaging method). The purpose of this study is to determine the degree of agreement between the imaging method and the palpation and visual methods (clinical method). In addition, the influence of differences in the information content of the sagittal plane images on the assessment was also investigated. In this experiment, 28 sagittal photographs of older adults whose posture had already been assessed using the clinical method were used. Furthermore, based on these photographs, 28 gray and silhouette images (G and S images) were generated, respectively. The G and S images were assessed by 28 physical therapists (PTs) using the imaging method. The assessment was based on the Kendall classification, with one of four categories selected for each image: ideal, kyphosis lordosis, sway back, and flat back. Cross-tabulation matrices of the assessments using the clinical method and imaging method were created. In this table, four categories and two categories of ideal and non-ideal (KL, SB, and FB) were created. The agreement was evaluated using the prevalence-adjusted bias-adjusted kappa (PABAK). In addition, sensitivity and specificity were calculated to confirm the reliability. When comparing the clinical and imaging methods in the four posture categories, the PABAK values were −0.14 and −0.29 for the S and G images, respectively. In the case of the two categories, the PABAK values were 0.57 and 0.5 for the S and G images, respectively. The sensitivity and specificity were 86% and 57% for the S images and 76% and 71% for the G images, respectively. The four categories show that the imaging method is difficult to assess regardless of the image processing. However, in the case of the two categories, the same assessment of the clinical method applied to the imaging method for both the S and G images. Therefore, no differences in image processing were observed, suggesting that PTs can identify posture using the visual method.

Barberi E., Chillemi M., Cucinotta F., Sfravara F.

Ergonomics focuses on the analysis of the interaction between human beings and their working environment. During the riding of a motorbike, ergonomics studies the rider’s posture on the motorbike. An incorrect posture can lead to physical and psychological discomfort, and can affect the perception of risk and the handling of the motorcycle. It is important for motorcyclists to adopt a good riding posture, for their health and road safety. The aim of this work is to propose a fast, cheap, and sufficiently robust method for the 3D reconstruction of the posture assumed by a motorcyclist. The stereo vision and the application of OpenPose made it possible to obtain a 3D reconstruction of the key points, and their evolution over time. The evaluation of the distances between the 3D key points, which represent the length of the various parts of the body, appears to remain sufficiently stable over time, and faithful to the real distances, as taken on the motorcyclist themself. The 3D reconstruction obtained can be applied in different fields: ergonomics, motorsport training, dynamics, and fluid dynamics analysis.

Kondou H., Morohashi R., Ichioka H., Bandou R., Matsunari R., Kawamoto M., Idota N., Ting D., Kimura S., Ikegaya H.

Although age estimation upon death is important in the identification of unknown cadavers for forensic scientists, to the best of our knowledge, no study has examined the utility of deep neural network (DNN) models for age estimation among cadavers. We performed a postmortem computed tomography (CT) examination of 1000 and 500 male and female cadavers, respectively. These CT slices were converted into 3-dimensional images, and only the thoracolumbar region was extracted. Eighty percent of them were categorized as training datasets and the others as test datasets for both sexes. We fine-tuned the ResNet152 models using the training datasets. We conducted 4-fold cross-validation, and the mean absolute error (MAE) of the test datasets was calculated using the ensemble learning of four ResNet152 models. Consequently, the MAE of the male and female models was 7.25 and 7.16, respectively. Our study shows that DNN models can be useful tools in the field of forensic medicine.

Zhang X., Fan J., Peng T., Zheng P., Zhang X., Tang R.

Recognizing sitting posture is significant to prevent the development of work-related musculoskeletal disorders for office workers. Multimodal data, i.e., infrared map and pressure map, have been leveraged to achieve accurate recognition while preserving privacy and being unobtrusive for daily use. Existing studies in sitting posture recognition utilize handcrafted features with machine learning models for multimodal data fusion, which significantly relies on domain knowledge. Therefore, a deep learning model is proposed to fuse the multimodal data and recognize the sitting posture. This model contains modality-specific backbones, a cross-modal self-attention module, and multi-task learning-based classification. Experiments are conducted to verify the effectiveness of the proposed model using 20 participants’ data, achieving a 93.08% F1-score. The high-performance result indicates that the proposed model is promising for sitting posture-related applications.

Kulkarni S., Deshmukh S., Fernandes F., Patil A., Jabade V.

Human pose estimation is the process of detecting the body keypoints of a person and can be used to classify different poses. Many researchers have proposed various ways to get a perfect 2D as well as a 3D human pose estimator that could be applied for various types of applications. This paper is a review of all the state-of-the-art architectures based on human pose estimation, the papers referred were based on the types of computer vision and machine learning algorithms, such as feed-forward neural networks, convolutional neural networks (CNN), OpenPose, MediaPipe, and many more. These different approaches are compared on various parameters, like the type of dataset used, the evaluation metric, etc. Different human pose datasets, such as COCO and MPII activity datasets with keypoints, as well as specific application-based datasets, are reviewed in this survey paper. Researchers may use these architectures and datasets in a range of domains, which are also discussed. The paper analyzes several approaches and architectures that can be used as a guide for other researchers to assist them in developing better techniques to achieve high accuracy.

Altameem A., Mahanty C., Poonia R.C., Saudagar A.K., Kumar R.

Breast cancer has evolved as the most lethal illness impacting women all over the globe. Breast cancer may be detected early, which reduces mortality and increases the chances of a full recovery. Researchers all around the world are working on breast cancer screening tools based on medical imaging. Deep learning approaches have piqued the attention of many in the medical imaging field due to their rapid growth. In this research, mammography pictures were utilized to detect breast cancer. We have used four mammography imaging datasets with a similar number of 1145 normal, benign, and malignant pictures using various deep CNN (Inception V4, ResNet-164, VGG-11, and DenseNet121) models as base classifiers. The proposed technique employs an ensemble approach in which the Gompertz function is used to build fuzzy rankings of the base classification techniques, and the decision scores of the base models are adaptively combined to construct final predictions. The proposed fuzzy ensemble techniques outperform each individual transfer learning methodology as well as multiple advanced ensemble strategies (Weighted Average, Sugeno Integral) with reference to prediction and accuracy. The suggested Inception V4 ensemble model with fuzzy rank based Gompertz function has a 99.32% accuracy rate. We believe that the suggested approach will be of tremendous value to healthcare practitioners in identifying breast cancer patients early on, perhaps leading to an immediate diagnosis.

Wang J., Chen D., Zhu M., Sun Y.

Construction workers are at high risk of work-related musculoskeletal disorders (WMSDs). Although the existing observational assessment methods are easy to use, when it comes to a more in-depth statistical analysis of the dynamic characteristics of the worker's operation, the sample data to be processed turn out to be large, the labor cost high, and the analysis easily affected by the prejudice of the evaluator. This study examines a novel WMSD prediction method based on the dynamic characteristics of the working posture, which comprises three artificial intelligence algorithms in series. In this method, the posture detector identifies the limb angles and state in the working video, the posture risk evaluator evaluates the risk level of the working posture frame by frame, and the task risk predictor predicts the risk level of the current work process. The collected video data of common tasks of construction workers and the MPII Human Pose dataset were used for training and evaluation of the algorithms. The method achieved 87.0% accuracy of the joint point recognition. The micro-averaged accuracy, recall, and F1-score (harmonic average of accuracy and recall) reached 96.7%, 96.0%, and 96.6%, respectively. The results showed that the proposed method has great potential for real-time risk assessment. It can output all of the changes of the limb angles of workers in the work process frame by frame and predict the risk level of the whole work process. • An automated method based on vision recognition algorithms is proposed to monitor, evaluate, and predict work posture risks. • The method is composed of three AI algorithms in series (posture detector, posture risk assessor, and task risk predictor). • The method achieved 87.0% accuracy of joint point recognition and greater than 96.0% accuracy of posture risk prediction. • The method can identify and evaluate the WMSD risk level of the work posture in a work video frame by frame. • The risk level of the work process is predicted according to the extracted posture risk change characteristics.

Piñero-Fuentes E., Canas-Moreno S., Rios-Navarro A., Domínguez-Morales M., Sevillano J.L., Linares-Barranco A.

The change from face-to-face work to teleworking caused by the pandemic has induced multiple workers to spend more time than usual in front of a computer; in addition, the sudden installation of workstations in homes means that not all of them meet the necessary characteristics for the worker to be able to position himself/herself comfortably with the correct posture in front of their computer. Furthermore, from the point of view of the medical personnel in charge of occupational risk prevention, an automated tool able to quantify the degree of incorrectness of a postural habit in a worker is needed. For this purpose, in this work, a system based on the postural detection of the worker is designed, implemented and tested, using a specialized hardware system that processes video in real time through convolutional neural networks. This system is capable of detecting the posture of the neck, shoulders and arms, providing recommendations to the worker in order to prevent possible health problems, due to poor posture. The results of the proposed system show that this video processing can be carried out in real time (up to 25 processed frames/sec) with a low power consumption (less than 10 watts) using specialized hardware, obtaining an accuracy of over 80% in terms of the pattern detected.

Kim W., Sung J., Saakes D., Huang C., Xiong S.

Observational ergonomic postural assessment methods have been commonly used to evaluate the risks of musculoskeletal disorders. Researchers have proposed semi-automatic methods using Kinect, known for limitations with body occlusions and non-frontal tracking. Meanwhile, new human pose estimation methods have been actively developed, and a popular open-source technology is OpenPose. This study aims to propose the OpenPose-based system for computing joint angles and RULA/REBA scores and validate against the reference motion capture system, and compare its performance to the Kinect-based system. Recordings of 10 participants performing 12 experimental tasks under different conditions: with/without body occlusions and tracked from frontal/non-frontal views were analyzed. OpenPose showed good performance under all task conditions, whereas Kinect performed significantly worse than OpenPose especially at cases with body occlusions or non-frontal tracking. The findings suggested that OpenPose could be a promising technology to measure joint angles and conduct semi-automatic ergonomic postural assessments in the real workspace where the conditions are often non-ideal. • This study proposes RULA/REBA ergonomic postural assessments using OpenPose. • Joint angles and RULA/REBA scores from OpenPose were validated against reference. • OpenPose-based ergonomic assessments were robust to non-ideal task conditions. • Kinect-based assessments were less accurate at non-frontal or occluded conditions. • OpenPose is promising for ergonomic postural assessments in the real workspace.

Maekawa M., Yoshizawa E., Hayata G., Ohashi S.

Poor posture has been shown to decrease both visceral and respiratory/circulatory function as well as to increase neuro-musculoskeletal system stress. Improper postures of children at school and in daily life can affect their physical and psychological development. In particular, many children who refuse to go to school or who have experienced school refusal have physical and mental problems. Given that posture is closely related to one’s psychological state, modifying one’s posture can improve both physical and psychological health problems. This study examined the changes to school-refusing students’ physical and psychological condition after attending an intervention to improve their posture. The participants were 65 high school students who have experienced school refusal and were attending a program to modify their posture for 2 months. Their posture and psychological states were recorded both pre- and post-intervention with the following measurements: postural alignment and spinal curve according to a sagittal plane, the General Health Questionnaire 30 (GHQ), and the Subjective Adjustment Scale (SAS). Post-intervention, most of the participants saw improvement in their postural alignment (e.g., trunk inclination in standing position, P < 0.001, 95%CI [2.00, 4.00]). Participants with improved standing posture post-intervention had higher SAS scores (e.g., feeling of acceptance and trust, P < 0.05, 95%CI [−3.00, −0.00]). We found that easy-to-implement postural interventions have a positive effect on students’ mental health. Furthermore, it was suggested that their adjustment to school would also improve as their posture improved. The contribution of this study shows that it is possible to care for the physical and mental health of students without using special facilities and techniques. It is hoped that the findings of this study will lead to an improved adjustment to both school or novel environments, as well as prevent health-based school refusal.

Alzubaidi L., Zhang J., Humaidi A.J., Al-Dujaili A., Duan Y., Al-Shamma O., Santamaría J., Fadhel M.A., Al-Amidie M., Farhan L.

In the last few years, the deep learning (DL) computing paradigm has been deemed the Gold Standard in the machine learning (ML) community. Moreover, it has gradually become the most widely used computational approach in the field of ML, thus achieving outstanding results on several complex cognitive tasks, matching or even beating those provided by human performance. One of the benefits of DL is the ability to learn massive amounts of data. The DL field has grown fast in the last few years and it has been extensively used to successfully address a wide range of traditional applications. More importantly, DL has outperformed well-known ML techniques in many domains, e.g., cybersecurity, natural language processing, bioinformatics, robotics and control, and medical information processing, among many others. Despite it has been contributed several works reviewing the State-of-the-Art on DL, all of them only tackled one aspect of the DL, which leads to an overall lack of knowledge about it. Therefore, in this contribution, we propose using a more holistic approach in order to provide a more suitable starting point from which to develop a full understanding of DL. Specifically, this review attempts to provide a more comprehensive survey of the most important aspects of DL and including those enhancements recently added to the field. In particular, this paper outlines the importance of DL, presents the types of DL techniques and networks. It then presents convolutional neural networks (CNNs) which the most utilized DL network type and describes the development of CNNs architectures together with their main features, e.g., starting with the AlexNet network and closing with the High-Resolution network (HR.Net). Finally, we further present the challenges and suggested solutions to help researchers understand the existing research gaps. It is followed by a list of the major DL applications. Computational tools including FPGA, GPU, and CPU are summarized along with a description of their influence on DL. The paper ends with the evolution matrix, benchmark datasets, and summary and conclusion.

Cao Z., Hidalgo G., Simon T., Wei S., Sheikh Y.

Realtime multi-person 2D pose estimation is a key component in enabling machines to have an understanding of people in images and videos. In this work, we present a realtime approach to detect the 2D pose of multiple people in an image. The proposed method uses a nonparametric representation, which we refer to as Part Affinity Fields (PAFs), to learn to associate body parts with individuals in the image. This bottom-up system achieves high accuracy and realtime performance, regardless of the number of people in the image. In previous work, PAFs and body part location estimation were refined simultaneously across training stages. We demonstrate that a PAF-only refinement rather than both PAF and body part location refinement results in a substantial increase in both runtime performance and accuracy. We also present the first combined body and foot keypoint detector, based on an internal annotated foot dataset that we have publicly released. We show that the combined detector not only reduces the inference time compared to running them sequentially, but also maintains the accuracy of each component individually. This work has culminated in the release of OpenPose, the first open-source realtime system for multi-person 2D pose detection, including body, foot, hand, and facial keypoints.

Song J., Gao S., Zhu Y., Ma C.

With the development of earth observation technologies, the acquired remote sensing images are increasing dramatically, and a new era of big data in remote sensing is coming. How to effectively min...