Movement recognition via channel-activation-wise sEMG attention.

CONTEXT: Surface electromyography (sEMG) signals contain rich information recorded from muscle movements and therefore reflect the user's intention. sEMG has seen dominant applications in rehabilitation, clinical diagnosis as well as human engineering, etc. However, current feature extraction methods for sEMG signals have been seriously limited by their stochasticity, transiency, and non-stationarity. OBJECTIVE: Our objective is to combat the difficulties induced by the aforementioned downsides of sEMG and thereby extract representative features for various downstream movement recognition. METHOD: We propose a novel 3-axis view of sEMG features composed of temporal, spatial, and channel-wise summary. We leverage the state-of-the-art architecture Transformer to enforce efficient parallel search and to get rid of limitations imposed by previous work in gesture classification. The transformer model is designed on top of an attention-based module, which allows for the extraction of global contextual relevance among channels and the use of this relevance for sEMG recognition. RESULTS: We compared the proposed method against existing methods on two Ninapro datasets consisting of data from both healthy people and amputees. Experimental results show the proposed method attains the state-of-the-art (SOTA) accuracy on both datasets. We further show that the proposed method enjoys strong generalization ability: a new SOTA is achieved by pretraining the model on a different dataset followed by fine-tuning it on the target dataset.

Stress Estimation Using Biometric and Activity Indicators to Improve QoL of the Elderly.

It is essential to estimate the stress state of the elderly to improve their QoL. Stress states change every day and hour, depending on the activities performed and the duration/intensity. However, most existing studies estimate stress states using only biometric information or specific activities (e.g., sleep duration, exercise duration/amount, etc.) as explanatory variables and do not consider all daily living activities. It is necessary to link various daily living activities and biometric information in order to estimate the stress state more accurately. Specifically, we construct a stress estimation model using machine learning with the answers to a stress status questionnaire obtained every morning and evening as the ground truth and the biometric data during each of the performed activities and the new proposed indicator including biological and activity perspectives as the features. We used the following methods: Baseline Method 1, in which the RRI variance and Lorenz plot area for 4 h after waking and 24 h before the questionnaire were used as features; Baseline Method 2, in which sleep time was added as a feature to Baseline Method 1; the proposed method, in which the Lorenz plot area per activity and total time per activity were added. We compared the results with the proposed method, which added the new indicators as the features. The results of the evaluation experiments using the one-month data collected from five elderly households showed that the proposed method had an average estimation accuracy of 59%, 7% better than Baseline Method 1 (52%) and 4% better than Baseline Method 2 (55%).

Estimating Congestion in a Fixed-Route Bus by Using BLE Signals.

Information on congestion of buses, which are one of the major public transportation modes, can be very useful in light of the current COVID-19 pandemic. Because it is unrealistic to manually monitor the number of riders on all buses in operation, a system that can automatically monitor congestion is necessary. The main goal of this paper's work is to automatically estimate the congestion level on a bus route with acceptable performance. For practical operation, it is necessary to design a system that does not infringe on the privacy of passengers and ensures the safety of passengers and the installation sites. In this paper, we propose a congestion estimation system that protects passengers' privacy and reduces the installation cost by using Bluetooth low-energy (BLE) signals as sensing data. The proposed system consists of (1) a sensing mechanism that acquires BLE signals emitted from passengers' mobile terminals in the bus and (2) a mechanism that estimates the degree of congestion in the bus from the data obtained by the sensing mechanism. To evaluate the effectiveness of the proposed system, we conducted a data collection experiment on an actual bus route in cooperation with Nara Kotsu Co., Ltd. The results showed that the proposed system could estimate the number of passengers with a mean absolute error of 2.49 passengers (error rate of 38.8%).

SALON: Simplified Sensing System for Activity of Daily Living in Ordinary Home.

As aging populations continue to grow, primarily in developed countries, there are increasing demands for the system that monitors the activities of elderly people while continuing to allow them to pursue their individual, healthy, and independent lifestyles. Therefore, it is required to develop the activity of daily living (ADL) sensing systems that are based on high-performance sensors and information technologies. However, most of the systems that have been proposed to date have only been investigated and/or evaluated in experimental environments. When considering the spread of such systems to typical homes inhabited by elderly people, it is clear that such sensing systems will need to meet the following five requirements: (1) be inexpensive; (2) provide robustness; (3) protect privacy; (4) be maintenance-free; and, (5) work with a simple user interface. In this paper, we propose a novel senior-friendly ADL sensing system that can fulfill these requirements. More specifically, we achieve an easy collection of ADL data from elderly people while using a proposed system that consists of a small number of inexpensive energy harvesting sensors and simple annotation buttons, without the need for privacy-invasive cameras or microphones. In order to evaluate the practicality of our proposed system, we installed it in ten typical homes with elderly residents and collected the ADL data over a two-month period. We then visualized the collected data and performed activity recognition using a long short-term memory (LSTM) model. From the collected results, we confirmed that our proposed system, which is inexpensive and non-invasive, can correctly collect resident ADL data and could recognize activities from the collected data with a high recall rate of 72.3% on average. This result shows a high potential of our proposed system for application to services for elderly people.

A self-management system for preventing hyperglycemia through blood glucose level prediction and nudge-based food amount reduction.

In recent years, the number of diabetic patients has been increasing rapidly all over the world. Diabetes cannot be completely cured once it develops, so it is important to prevent diabetes before it develops. To prevent the onset of diabetes, it is necessary to avoid prolonged hyperglycemia after meals. In this paper, we propose a self-management system to help users prevent diabetes. The system monitors blood glucose levels in real time, calculates foods to be reduced taking into account the user's preferences, and presents them to the user as soon as the system predicts that the planned diet will cause high blood glucose. We designed and conducted two experiments to show the effectiveness of the proposed system. Experiment 1 was to construct and evaluate a model for predicting blood glucose levels two hours later. Experiment 2 was to evaluate the degree of satisfaction with the food and recommendations, and the acceptability of the recommendations by participants who actually used the proposed system. The results of Experiment 1 showed that the constructed model was able to predict blood glucose levels with an RMSE of 7.66 and MAE of 4.66. As a result of Experiment 2, we found the recommended intake was more acceptable if it reflected the user's preferences.

Influence of cortical bone thickness and implant length on implant stability at the time of surgery--clinical, prospective, biomechanical, and imaging study.

This clinical study is the first to quantitatively evaluate both regional bone structure by computed tomography preoperatively and dental implant stability by resonance frequency analysis at the time of surgery to explore the relation between local bone structure and dental implant stability in humans. Implant stability at the time of installation is often difficult to achieve in lower density bone and implant stability might influence treatment efficacy. Few clinical studies have reported detailed bone characteristics obtained using computed tomography prior to surgery and comprehensive implant stability measurements at the time of surgery. We hypothesized that thicker cortical bone would improve the stability of the dental implant at the time of placement. Before radiographic examination, diagnostic radiographic templates were made by incorporating radiopaque indicators. Computed tomography scans were obtained for 50 edentulous subjects prior to surgery. Preoperatively, the thickness of the cortical bone at the sites of implant insertion was measured digitally, and then implant insertion surgery was performed. A total of 225-implant stability measurements were made using a resonance frequency analyzer. There was a strong linear correlation between cortical bone thickness and resonance frequency (r = 0.84, P < 0.0001). The implant length had a weak negative correlation with stability (r = -0.25, P < 0.0005). These results suggest that the initial stability at the time of implant installation is influenced more by cortical bone thickness than by implant length. The cortical and cancellous ratio of local bone is extremely important for implant stability at the time of surgery and determining the local bone condition is critical for treatment success.

Pthlh, a promising cancer modifier gene in rat tongue carcinogenesis.

Susceptibly to the induction of rat tongue cancer (TC) by oral 4-nitroquinoline 1-oxide (4NQO) exposure is a polygenic trait. Among several quantitative trait loci identified by crosses between TC-susceptible Dark Agouti (DA) rats and TC-resistant Wistar-Furth (WF) rats, we focused on tongue cancer susceptibility locus (Tcas3) of chromosome 4. We examined tongue carcinogenesis in the reciprocal congenic strains DA.WF-Tcas3 and WF.DA-Tcas3 and in their parental strains. The Tcas3DA allele, and not the Tcas3WF allele, significantly favored tumor latency, incidence and TC number/size. In genomic DNA of TCs induced in (DA x WF) F1 rats, the resistant Tcas3WF allele was frequently and selectively lost, particularly in larger tumors. Thus, we searched the possible candidate genes in the Tcas3 region using microarray analysis of TCs in F1 rats and revealed significant upregulation of 2 cancer-related genes, parathyroid hormone-like hormone (Pthlh) and Kras2. The relevance of the WF allele of Pthlh as a cancer modifier was indicated by 3 single nucleotide polymorphisms specific to this strain. In contrast, no consistent strain-specific variations were found in Kras2. Moreover, the plasma Ca2+ level was consistently higher in DA rats when compared to the level in WF rats bearing TCs; moreover, the Pthlh-mRNA expression level was >30-fold higher in TCs when compared to this level in the normal tongue mucosa. Immunostaining experiments showed strong PTHrP protein expression in TCs of DA rats, and the signal was intensified in larger TCs. Kras2 was also upregulated in TCs, but to a lesser degree than PTHrP. Thus, Pthlh is a promising candidate modifier gene in the development and progression of rat TCs.

An emergency medical communications system by low altitude platform at the early stages of a natural disaster in Indonesia.

A natural disaster is a consequence of a natural hazard, such as a tsunami, earthquake or volcanic eruption, affecting humans. In order to support emergency medical communication services in natural disaster areas where the telecommunications facility has been seriously damaged, an ad hoc communication network backbone should be build to support emergency medical services. Combinations of requirements need to be considered before deciding on the best option. In the present study we have proposed a Low Altitude Platform consisting of tethered balloons combined with Wireless Fidelity (WiFi) 802.11 technology. To confirm that the suggested network would satisfy the emergency medical service requirements, a communications experiment, including performance service measurement, was carried out.

Development traumatic brain injury computer user interface for disaster area in Indonesia supported by emergency broadband access network.

Disasters bring consequences of negative impacts on the environment and human life. One of the common cause of critical condition is traumatic brain injury (TBI), namely, epidural (EDH) and subdural hematoma (SDH), due to downfall hard things during earthquake. We proposed and analyzed the user response, namely neurosurgeon, general doctor/surgeon and nurse when they interacted with TBI computer interface. The communication systems was supported by TBI web based applications using emergency broadband access network with tethered balloon and simulated in the field trial to evaluate the coverage area. The interface consisted of demography data and multi tabs for anamnesis, treatment, follow up and teleconference interfaces. The interface allows neurosurgeon, surgeon/general doctors and nurses to entry the EDH and SDH patient's data during referring them on the emergency simulation and evaluated based on time needs and their understanding. The average time needed was obtained after simulated by Lenovo T500 notebook using mouse; 8-10 min for neurosurgeons, 12-15 min for surgeons/general doctors and 15-19 min for nurses. By using Think Pad X201 Tablet, the time needed for entry data was 5-7 min for neurosurgeon, 7-10 min for surgeons/general doctors and 12-16 min for nurses. We observed that the time difference was depending on the computer type and user literacy qualification as well as their understanding on traumatic brain injury, particularly for the nurses. In conclusion, there are five data classification for simply TBI GUI, namely, 1) demography, 2) specific anamnesis for EDH and SDH, 3) treatment action and medicine of TBI, 4) follow up data display and 5) teleneurosurgery for streaming video consultation. The type of computer, particularly tablet PC was more convenient and faster for entry data, compare to that computer mouse touched pad. Emergency broadband access network using tethered balloon is possible to be employed to cover the communications systems in disaster area.

Characteristics and risk factors for typhoid fever after the tsunami, earthquake and under normal conditions in Indonesia.

BACKGROUND: Although typhoid transmitted by food and water is a common problem in daily life, its characteristics and risk factors may differ in disaster-affected areas, which reinforces the need for rapid public health intervention. Surveys were carried out post-tsunami in Banda Aceh, post-earthquake in Yogyakarta, and under normal conditions in Bandung, Indonesia. Logistic regression analysis was used to assess the risk factors with the dependent variable of typhoid fever, with or without complications. FINDINGS: Characteristic typhoid fever with complications was found in 5 patients (11.9%) affected by the tsunami in Aceh, 8 (20.5%) after the earthquake in Yogyakarta, and 13 (18.6%) in Bandung. After the tsunami in Aceh, clean water (OR = 0.05; 95%CI: 0.01-0.47) and drug availability (OR = 0.23; 95%CI: 0.02-2.43) are significant independent risk factors, while for the earthquake in Yogyakarta, contact with other typhoid patients (OR = 20.30; 95%CI: 1.93-213.02) and education (OR = 0.08; 95%CI: 0.01-0.98) were significant risk factors. Under normal conditions in Bandung, hand washing (OR = 0.07; 95%CI: 0.01-0.50) and education (OR = 0.08; 95%CI: 0.01-0.64) emerged as significant risk factors. CONCLUSION: The change in risk factors for typhoid complication after the tsunami in Aceh and the earthquake in Yogyakarta emphasizes the need for rapid public health intervention in natural disasters in Indonesia.

Designing an emergency medical information system for the early stages of disasters in developing countries: the human interface advantage, simplicity and efficiency.

The means of designing an effective user interface software package varies from one application to another. Almost the entire ICT infrastructure was damaged following the impact of the tsunami tidal wave. Under such circumstances, transporting critically ill patients is a must and becomes the first priority. Many considerations are needed when designing a specific user interface for emergency situations in developing countries. In this study we proposed how to design the user interface in order to support emergency medical care in the early stages of disasters. The user interface was classified into two tabs, firstly to indicate critically ill patients and secondly to notify details of the medicine having been administered to the patients. Classifying the user interface of emergency medical care information systems by using VHF radio connections will be beneficial, especially for the early stages of disaster-stricken developing countries, in order to preserve the lives of more victims.

Characteristic tetanus infection in disaster-affected areas: case study of the Yogyakarta earthquakes in Indonesia.

BACKGROUND: Tetanus is an infectious disease caused by the contamination of wounds from bacteria that live in soil. The tetanus mortality rate remains high in developing countries affected by natural disasters. Whether the socio-demography and geographical conditions may influence the tetanus treatment outcome on the earthquake situation in Yogyakarta, Indonesia has not been investigated. FINDINGS: We present 26 tetanus patients who were admitted to eight hospitals following the earthquakes that occurred on May, 27, 2006, in Yogyakarta, Indonesia. The independent variables were age, gender, distance, admission, hospitalization, and type of hospital with the dependent variable surviving or perishing. Data were analyzed by logistic regression methods on SPSS 17.0. The distance from the patient's place of residence to the hospital were obtained and analyzed by using geospatial tools MapInfo 7.8 SCP and Global Mapper 7. Eight of the 26 patients were dead (30.8%) and statistical results showed that the distance (OR = 1.740, 95% CI = 1.068-2.835) and type of hospital (OR = 0.067, 95% CI = 0.001-3.520) were significant predictors of death. CONCLUSION: Our findings show that in order to reduce the mortality rates, performing triage systems based on the distance and type of hospital priority for internally displaced persons could be proposed as well as making provisions for the generally old population in order to prevent an outbreak of tetanus following earthquakes in Yogyakarta, Indonesia.