Development of a Non-Contact Sensor System for Converting 2D Images into 3D Body Data: A Deep Learning Approach to Monitor Obesity and Body Shape in Individuals in Their 20s and 30s.

This study demonstrates how to generate a three-dimensional (3D) body model through a small number of images and derive body values similar to the actual values using generated 3D body data. In this study, a 3D body model that can be used for body type diagnosis was developed using two full-body pictures of the front and side taken with a mobile phone. For data training, 400 3D body datasets (male: 200, female: 200) provided by Size Korea were used, and four models, i.e., 3D recurrent reconstruction neural network, point cloud generative adversarial network, skinned multi-person linear model, and pixel-aligned impact function for high-resolution 3D human digitization, were used. The models proposed in this study were analyzed and compared. A total of 10 men and women were analyzed, and their corresponding 3D models were verified by comparing 3D body data derived from 2D image inputs with those obtained using a body scanner. The model was verified through the difference between 3D data derived from the 2D image and those derived using an actual body scanner. Unlike the 3D generation models that could not be used to derive the body values in this study, the proposed model was successfully used to derive various body values, indicating that this model can be implemented to identify various body types and monitor obesity in the future.

Establish a machine learning based model for optimal casting conditions management of small and medium sized die casting manufacturers.

Die casting is a suitable process for producing complex and high precision parts, but it faces challenges in terms of quality degradation due to inevitable defects. The casting parameters play a significant role in quality, and in many cases, producers rely on their experience to manage these parameters. In order to address this, domestic small and medium sized die casting companies have established smart factories (MES) and collected data. This study aims to utilize this data to construct a machine learning based optimal casting parameter model to enhance quality. During the model development process, distinct important features were identified for each company. This indicates the necessity of deriving tailored models for each site, aligning with the make to order (MTO) environment, rather than a generalized model.

Machine learning-based obesity classification considering 3D body scanner measurements.

Obesity can cause various diseases and is a serious health concern. BMI, which is currently the popular measure for judging obesity, does not accurately classify obesity; it reflects the height and weight but ignores the characteristics of an individual's body type. In order to overcome the limitations of classifying obesity using BMI, we considered 3-dimensional (3D) measurements of the human body. The scope of our study was limited to Korean subjects. In order to expand 3D body scan data clinically, 3D body scans, Dual-energy X-ray absorptiometry, and Bioelectrical Impedance Analysis data was collected pairwise for 160 Korean subjects. A machine learning-based obesity classification framework using 3D body scan data was designed, validated through Accuracy, Recall, Precision, and F1 score, and compared with BMI and BIA. In a test dataset of 40 people, BMI had the following values: Accuracy: 0.529, Recall: 0.472, Precision: 0.458, and F1 score: 0.462, while BIA had the following values: Accuracy: 0.752, Recall: 0.742, Precision: 0.751, and F1 score: 0.739. Our proposed model had the following values: Accuracy: 0.800, Recall: 0.767, Precision: 0.842, and F1 score: 0.792. Thus, our accuracy was higher than BMI as well as BIA. Our model can be used for obesity management through 3D body scans.

Development of an intrinsic health risk prediction model for camera-based monitoring of older adults living alone.

The number of older adults in Korea is increasing, along with the number of depressed older patients. The causes of depression in older adults include social isolation with negligible interaction with others, irregular nutritional habits, and self-negligence, i.e., they do not engage in any activity. These factors, self-negligence, social isolation, and irregular nutritional habits, are defined as inherent health risks, and in this study, we detected them. These factors can only be derived through long-term monitoring, but the current monitoring system for older adults is severely limited as it focuses only on emergencies, such as "falls." Therefore, in this study, the goal was to perform long-term monitoring using a camera. In order to capture the physical characteristics of the older adults, the ETRI-Activity3D data were used for training, and the skeleton-based action recognition algorithm Posec3d was used. By defining 90 frames as the time taken for one action, we built a monitoring system to enable long-term monitoring of older adult by performing multiple action detection in one video. A reliable monitoring system, with 98% accuracy, 98% precision, 99% recall, and 98% F1, was successfully established for health monitoring of older adults. This older adult monitoring technology is expected to improve the quality of medical services in a medical environment as well as the objective, activities of daily living test, which does not depend on the observer through daily life detection.

Development of an Obesity Information Diagnosis Model Reflecting Body Type Information Using 3D Body Information Values.

This study uses various body values (length, circumference, and volume) that can be derived from 3D data to determine variables and areas that substantially affect obesity and suggests guidelines for diagnosing obesity that are more elaborate than existing obesity indices. Body data for 170 participants (87 men and 73 women aged 20-30 years) are collected for the chest, abdomen, hips, and arms/legs. A 3D scanner, which can produce accurate body point results, and dual-energy X-ray (DEXA), which can accurately determine the fat percentage, are used to derive fat rates for each body part. The fat percentage and total fat percentage for each body part are used as learning data. For the derived data, the eigenvalue for each body part is derived using a principal component analysis, and the following four clusters are created for each part: underweight, normal, overweight, and obese. A comparison with the obesity index, which diagnoses obesity based on the cluster model, showed that the accuracy of the model proposed in this study is higher at 80%. Therefore, this model can determine the body information necessary for accurate obesity diagnosis and be used to diagnose obesity with greater accuracy than obesity indices without a body fat measurement machine such as DEXA.

Artificial intelligence in positioning between mandibular third molar and inferior alveolar nerve on panoramic radiography.

Determining the exact positional relationship between mandibular third molar (M3) and inferior alveolar nerve (IAN) is important for surgical extractions. Panoramic radiography is the most common dental imaging test. The purposes of this study were to develop an artificial intelligence (AI) model to determine two positional relationships (true contact and bucco-lingual position) between M3 and IAN when they were overlapped in panoramic radiographs and compare its performance with that of oral and maxillofacial surgery (OMFS) specialists. A total of 571 panoramic images of M3 from 394 patients was used for this study. Among the images, 202 were classified as true contact, 246 as intimate, 61 as IAN buccal position, and 62 as IAN lingual position. A deep convolutional neural network model with ResNet-50 architecture was trained for each task. We randomly split the dataset into 75% for training and validation and 25% for testing. Model performance was superior in bucco-lingual position determination (accuracy 0.76, precision 0.83, recall 0.67, and F1 score 0.73) to true contact position determination (accuracy 0.63, precision 0.62, recall 0.63, and F1 score 0.61). AI exhibited much higher accuracy in both position determinations compared to OMFS specialists. In determining true contact position, OMFS specialists demonstrated an accuracy of 52.68% to 69.64%, while the AI showed an accuracy of 72.32%. In determining bucco-lingual position, OMFS specialists showed an accuracy of 32.26% to 48.39%, and the AI showed an accuracy of 80.65%. Moreover, Cohen's kappa exhibited a substantial level of agreement for the AI (0.61) and poor agreements for OMFS specialists in bucco-lingual position determination. Determining the position relationship between M3 and IAN is possible using AI, especially in bucco-lingual positioning. The model could be used to support clinicians in the decision-making process for M3 treatment.

Development of a Wearable Camera and AI Algorithm for Medication Behavior Recognition.

As many as 40% to 50% of patients do not adhere to long-term medications for managing chronic conditions, such as diabetes or hypertension. Limited opportunity for medication monitoring is a major problem from the perspective of health professionals. The availability of prompt medication error reports can enable health professionals to provide immediate interventions for patients. Furthermore, it can enable clinical researchers to modify experiments easily and predict health levels based on medication compliance. This study proposes a method in which videos of patients taking medications are recorded using a camera image sensor integrated into a wearable device. The collected data are used as a training dataset based on applying the latest convolutional neural network (CNN) technique. As for an artificial intelligence (AI) algorithm to analyze the medication behavior, we constructed an object detection model (Model 1) using the faster region-based CNN technique and a second model that uses the combined feature values to perform action recognition (Model 2). Moreover, 50,000 image data were collected from 89 participants, and labeling was performed on different data categories to train the algorithm. The experimental combination of the object detection model (Model 1) and action recognition model (Model 2) was newly developed, and the accuracy was 92.7%, which is significantly high for medication behavior recognition. This study is expected to enable rapid intervention for providers seeking to treat patients through rapid reporting of drug errors.

Performance Prediction of a MongoDB-Based Traceability System in Smart Factory Supply Chains.

In the future, with the advent of the smart factory era, manufacturing and logistics processes will become more complex, and the complexity and criticality of traceability will further increase. This research aims at developing a performance assessment method to verify scalability when implementing traceability systems based on key technologies for smart factories, such as Internet of Things (IoT) and BigData. To this end, based on existing research, we analyzed traceability requirements and an event schema for storing traceability data in MongoDB, a document-based Not Only SQL (NoSQL) database. Next, we analyzed the algorithm of the most representative traceability query and defined a query-level performance model, which is composed of response times for the components of the traceability query algorithm. Next, this performance model was solidified as a linear regression model because the response times increase linearly by a benchmark test. Finally, for a case analysis, we applied the performance model to a virtual automobile parts logistics. As a result of the case study, we verified the scalability of a MongoDB-based traceability system and predicted the point when data node servers should be expanded in this case. The traceability system performance assessment method proposed in this research can be used as a decision-making tool for hardware capacity planning during the initial stage of construction of traceability systems and during their operational phase.

Environment and Its Influence on Health and Demographics in South Korea.

As the prevalence of overweight and obesity has been increasing in South Korea, it is critical to better understand possible associations between environmental surroundings and general health status. We characterize key health test readings and basic demographic information from 10,816 South Koreans, obtained from two Ubiquitous Healthcare (U-Healthcare) centers that have distinct surrounding neighborhood characteristics. One is located in a rural area in Busan, the other is located in an urban area in Daegu surrounded by a highly crowded residential and commercial business area. We analyze comprehensive health data sets, including blood pressure, body mass index, pulse rate, and body fat percentage from December 2013 to December 2014 to study differences in overall health test measurements between users of rural and urban U-Healthcare centers. We conduct multiple regression analyses to evaluate differences in general health status between the two centers, adjusting for confounding factors. We report statistical evidence of differences in blood pressure at the two locations. As local residents are major users, the result indicates that the environmental surroundings of the centers can influence the demographics of the users, the type of health tests in demand, and the users' health status. We further envision that U-Healthcare centers will provide public users with an opportunity for enhancing their current health, which could potentially be used to prevent them from developing chronic diseases, while providing surveillance healthcare data.

Wearable Sensing of In-Ear Pressure for Heart Rate Monitoring with a Piezoelectric Sensor.

In this study, we developed a novel heart rate (HR) monitoring approach in which we measure the pressure variance of the surface of the ear canal. A scissor-shaped apparatus equipped with a piezoelectric film sensor and a hardware circuit module was designed for high wearability and to obtain stable measurement. In the proposed device, the film sensor converts in-ear pulse waves (EPW) into electrical current, and the circuit module enhances the EPW and suppresses noise. A real-time algorithm embedded in the circuit module performs morphological conversions to make the EPW more distinct and knowledge-based rules are used to detect EPW peaks. In a clinical experiment conducted using a reference electrocardiogram (ECG) device, EPW and ECG were concurrently recorded from 58 healthy subjects. The EPW intervals between successive peaks and their corresponding ECG intervals were then compared to each other. Promising results were obtained from the samples, specifically, a sensitivity of 97.25%, positive predictive value of 97.17%, and mean absolute difference of 0.62. Thus, highly accurate HR was obtained from in-ear pressure variance. Consequently, we believe that our proposed approach could be used to monitor vital signs and also utilized in diverse applications in the near future.

U-Healthcare Center Service in Busan City, South Korea: An Empirical Analysis and the Results of 1 Year of Service.

INTRODUCTION: Studies have demonstrated that technological innovation is vital for prosperous economies, and greater technological innovation leads to improved public health indicators. The South Korean government has implemented policies to provide city services using information communication technologies, and ubiquitous healthcare (u-healthcare) wellness is one of these. This article presents the effects of using a u-healthcare center model that proves self-healthcare monitoring can work for the general population. MATERIALS AND METHODS: The u-healthcare center has provided service to the public since April 2013. It is equipped with medical devices that evaluate physiological parameters such as weight, body mass index (BMI), blood pressure (BP), pulse rate (PR), and body fat (BF). This article focuses on the analysis of BMI, BP, PR, and BF parameters. RESULTS: Health test results from 12,766 voluntary patients of the u-healthcare center were analyzed during a 1-year period. The four health parameters from each of the four seasons were analyzed and compared, showing statistically significant seasonal differences. A Duncan's post hoc analysis showed that BMI did not differ between spring and summer, whereas BP differed throughout all seasons. Participation of females was higher compared with males, and men's average BMI was statistically higher than that of the women. Some additional significant findings for all participants were as follows: 48.8% scored normal in BMI, 31.7% scored normal-controlled in BP, 90.7% scored normal in PR, and 24.8% scored normal in BF. A survey showed that 96.4% found the u-healthcare center to be generally helpful, and 95.7% responded that they would recommend it. CONCLUSIONS: Implementation of u-healthcare projects provides a new public service toward evaluating health parameters, providing historical health information access, promoting self-monitoring, and motivating users to be more aware of their own health status.

The Development and User Satisfaction Evaluation of Internet-Based N-Screen Healthcare Walking Content to Increase Continuous Usage Motivation.

OBJECTIVE: The purpose of the current study is (1) to apply Internet-based N-Screen (this is used like the term "emultiscreen"; as the technology that provides services of shared content or application via N devices, it includes all screens such as personal computers [PCs], TV, and mobile devices) services to healthcare services by developing games for improving one's health and (2) to present ways to activate the use of health promotion contents in the future by investigating user satisfaction and whether there is any intention to accept the contents and/or use the services continuously. MATERIALS AND METHODS: In order to evaluate the customized health maintenance content provided by the healthcare walking system developed in the current study, 98 adult men and women residing in Seoul, Korea, were instructed to use 10 minutes' worth of the walking content. Perceived quality, level of trust in the results, effectiveness of the exercise, and overall satisfaction were measured in regard to the N-Screen-based walking content, including those for the cell phone, PC, and Internet protocol TV (IPTV). RESULTS: Walking contents using N-Screen services were perceived with high levels of trust in the results of the exercise, the effectiveness of the exercise, and overall satisfaction. In terms of the usability of N-Screen services, the younger the participants, the more usable they found the mobile or PC programs. The older the participants, the more usable they found the IPTV screens, although they still struggled with using the content given; operating IPTVs proved to be difficult for them. Furthermore, participants who were engaged in exercise on a regular basis were less satisfied with the program, in general. CONCLUSIONS: The present study has developed a walking system using N-Screen programs to make the most common and effective forms of exercise-walking and running-accessible indoors. This may increase motivation to exercise by offering services that boost one's interest in exercising, such as personal monitoring and real-time feedback regarding one's workout progress.

How the awareness of u-healthcare service and health conditions affect healthy lifestyle: an empirical analysis based on a u-healthcare service experience.

OBJECTIVE: The objectives of this study are (1) to establish a ubiquitous healthcare (u-healthcare) center for those who wish to use u-healthcare, allowing them to experience the service, and (2) to evaluate the users' awareness and expectations of the service based on their overall assessment. MATERIALS AND METHODS: To establish the u-healthcare center, a kiosk, devices for health checkup, a body-type examination system, and a physical fitness assessment system were installed. Also, a u-healthcare Web site was developed. A survey was conducted on 280 individuals who visited the u-healthcare center and used the service, to determine (1) individual awareness of u-healthcare before using the service and their change of perception after use, (2) factors that affect the use of u-healthcare, and (3) the effects of disease awareness on exercise habits. RESULTS: Only 25.4% of the participants were aware of u-healthcare, and only 36% who saw the u-healthcare center recognized that it was where the u-healthcare service was provided. The group of individuals who were willing to use the u-healthcare showed statistically significant differences in their satisfaction with the overall environment of the center, as well as the specificity of the descriptions, examination results, kindness of the staff, and their responses. Additionally, the group of individuals who were diagnosed with chronic diseases and the group who were not showed statistically significant differences in the number of days on which they exercised lightly or took a walk. CONCLUSIONS: To promote the usage of u-healthcare service, the understanding of the service and the credibility of examination results need to be increased by sharing successful cases. Furthermore, to expand the use of the system that allows a person to regularly check his or her state of health, a lifelong periodical management system linked with another medical welfare program will be needed.

Development and evaluation of a mobile application for personal lifestyle check-up and improvement.

OBJECTIVE: This study aimed (1) to help individuals analyze their own health status by checking their lifestyle, (2) to develop a user-friendly mobile application that offered prescriptions for lifestyle improvement, and (3) to examine whether the developed application had positive effects on users. MATERIALS AND METHODS: In order to develop a lifestyle analysis engine that would operate in an Android(®) (Google, Mountain View, CA)-based mobile application, survey data on health awareness behaviors of 25,124 participants from the 2009 Korean National Health and Nutrition Examination Survey (KNHANES) were analyzed. Additionally, in order for the users to be aware of their lifestyles and explore the effects of the developed mobile application on lifestyle management and improvement, an additional survey of the lifestyle awareness and levels of motivation for lifestyle improvement of 152 users was conducted. RESULTS: The differences between lifestyles before and after using the application were examined. A paired t test was used for questions regarding (1) the level of motivation to improve lifestyles and (2) changes in lifestyle. The lifestyle score was lower after using the program than before using it. Conversely, the level of motivation to improve lifestyle was greater after the program than before it. Both results were statistically significant. CONCLUSIONS: By using the KNHANES, this study developed a mobile application that compared the quantified lifestyles of individuals and enabled individuals to check easily their health statuses, whenever and wherever necessary. The program developed in this study contributed to motivating individuals to be aware of and to improve their lifestyles.