In recent years, the declining birthrate and aging population have become serious problems in Japan. To solve these problems, we have developed a system based on edge AI. This system predicts the future heart rate during walking in real time and provides feedback to improve the quality of exercise and extend healthy life expectancy. In this paper, we predicted the heart rate in real time based on the proposed system and provided feedback. Experiments were conducted without and with the predicted heart rate, and a comparison was made to demonstrate the effectiveness of the predicted heart rate.

In this study, we propose a method to estimate body composition-related health indicators (e.g., ratio of body fat, body water, and muscle, etc.) using video-based gait analysis. This method is more efficient than individual measurement using a conventional body composition meter. Specifically, we designed a deep-learning framework with a convolutional neural network (CNN), where the input is a gait energy image (GEI) and the output consists of the health indicators. Although a vast amount of training data is typically required to train network parameters, it is unfeasible to collect sufficient ground-truth data, i.e., pairs consisting of the gait video and the health indicators measured using a body composition meter for each subject. We therefore use a two-step approach to exploit an auxiliary gait dataset that contains a large number of subjects but lacks the ground-truth health indicators. At the first step, we pre-train a backbone network using the auxiliary dataset to output gait primitives such as arm swing, stride, the degree of stoop, and the body width — considered to be relevant to the health indicators. At the second step, we add some layers to the backbone network and fine-tune the entire network to output the health indicators even with a limited number of ground-truth data points of the health indicators. Experimental results show that the proposed method outperforms the other methods when training from scratch as well as when using an auto-encoder-based pre-training and fine-tuning approach; it achieves relatively high estimation accuracy for the body composition-related health indicators except for body fat-relevant ones.

This paper provides perspectives for future medical healthcare social services and businesses that integrate advanced information and communication technology (ICT) and data science. First, we propose a universal medical healthcare platform that consists of wireless body area network (BAN), cloud network and edge computer, big data mining server and repository with machine learning. Technical aspects of the platform are discussed, including the requirements of reliability, safety and security, i.e., so-called dependability. In addition, novel technologies for satisfying the requirements are introduced. Then primary uses of the platform for personalized medicine and regulatory compliance, and its secondary uses for commercial business and sustainable operation are discussed. We are aiming at operate the universal medical healthcare platform, which is based on the principle of regulatory science, regionally and globally. In this paper, trials carried out in Kanagawa, Japan and Oulu, Finland will be revealed to illustrate a future medical healthcare social infrastructure by expanding it to Asia-Pacific, Europe and the rest of the world. We are representing the activities of Kanagawa medical device regulatory science center and a joint proposal on security in the dependable medical healthcare platform. Novel schemes of ubiquitous rehabilitation based on analyses of the training effect by remote monitoring of activities and machine learning of patient's electrocardiography (ECG) with a neural network are proposed and briefly investigated.

This paper surveys advances in biomedical processor SoC technology for healthcare application and reviews state-of-the-art architecture and circuits used in SoC integration. Particularly, this paper categorizes and describes techniques for improving power efficiency in communication, computation, and sensing. Additionally, it surveys accuracy enhancement techniques for bio-signal measurement and recognition. Finally, we have discussed the potential new directions for development as well as research.

This study proposes a design and calibration method for a small-footprint, low-frequency, and low-power gate leakage timer using a differential leakage technique for IoT applications. The proposed gate leakage timer is different from conventional ones because it is composed of two leakage sources and exploits differential leakage current for the charging capacitor. This solution alleviates the inherent trade-off between small-footprint and low-frequency in the conventional gate leakage timer. Furthermore, a calibration method to suppress variations of the output frequency is proposed in this paper. To verify the effectiveness of the proposed gate leakage timer, a test chip was fabricated using 55-nm-DDC-CMOS technology. The test chip successfully demonstrates the highest figure of merit (FoM) of the product of the capacitor area (0.072µm2) and output frequency (0.11Hz), which corresponds to an improvement by a factor of 2,121 compared to the conventional one. It also demonstrates the operation with 4.5pW power consumption. The total footprint can be reduced to be 28µm2, which enables low-cost and low-power IoT edges. The scaling scenario shows that the proposed technique is conducive to technology scaling.

With the rapid merger of healthcare business and information technology, more healthcare institutions and medical practices are sharing information. Since these records often contain patients' sensitive personal information, Healthcare Information Systems (HISs) should be properly designed to manage these records in a secure manner. We propose a novel security design for the HIS complying with the security and privacy rules. The proposed system defines protocols to ensure secure delivery of medical records over insecure public networks and reliable management of medical record in the remote server without incurring excessive costs to implement services for security. We demonstrate the practicality of the proposed system through a security analysis and performance evaluation.

Infrastructures for the evaluation of the state of health of individuals using a standardized communication network consisting of advanced instruments and subsequent data analysis have been developed. Here we report that this developed infrastructure has been tested in the field in 100 houses and involving almost 300 users. The communication protocol part of this infrastructure has been standardized as IEEE 11073-20601. Continua Health Alliance, an international not-for-profit industry organization which has nearly 230 member companies, has adopted this IEEE 11073-20601 to establish an ecosystem of interoperable personal connected health systems that empower individuals and organizations to better manage their health and wellness. Currently nearly 100 Continua certified products are available in public including smartphone.

In this paper, we propose a jogging support system that plays back background music while synchronizing its tempo with the user's jogging pace. Keeping an even pace is important in jogging but it is not easy due to tiredness. Our system indicates the variation of the runner's pace by changing the playback speed of music according to the user's pace variation. Because this requires the runner to keep an even pace in order to enjoy the music at its normal speed, the runner will be spontaneously influenced to keep an even pace. Experimental results show that our system reduced the variation of jogging pace.

We propose a system to provide accurate data and timely medical services, unconstrained by location, through the use of a platform that can utilize mobile devices, interface with sensors and medical information systems. As the application of integrated platform, we aim to develop medical services that manage thrombolytic therapy for emergent stroke patients.

Various information and communication technology (ICT) applications have been introduced for use in disaster-stricken areas. Of these, those for healthcare purposes proved useful when used for continually monitoring health conditions and easily using medical and healthcare devices, such as sphygmomanometers, which are familiar to the public. Devices that were easy to understand and use were more acceptable in the first post-disaster stage in a temporary housing community as victims were experiencing mental and physical trauma. After gaining the trust of medical/healthcare staff caring for the victims, we could move to the next stage with their support. ICT technologies are tools and are basically invisible to victims; a notion we have recognized anew. This paper introduces our activities in the area of healthcare monitoring services. The UMe-1 system and subset health check kiosk version 0 are introduced. The kiosk is simple and important for life support advisers in support centers as a tool for checking the health condition of residents. Not all those living in the target temporary housing community are currently using the kiosk due to its location within the widespread premises, but the director of a support center recognizes the utility of the kiosk and expects that a community-based kiosk and user-friendly BAN device could be used at home for mimamori (the Japanese concept of informally monitoring healthcare) because residents living far from support centers have difficulty regularly visiting the center, especially in the winter. We introduced a television-based, in-home health check kiosk and are currently working on its development. There are gaps between actually deployable technologies and research results, but it is also important to continue to address research issues concerning reducing impact through the user interface and introducing less stress to the everyday lives of disaster victims.

Large-scale adoption of electronic healthcare applications requires semantic interoperability. The new proposals propose an advanced (multi-level) DBMS architecture for repository services for health records of patients. These also require query interfaces at multiple levels and at the level of semi-skilled users. In this regard, a high-level user interface for querying the new form of standardized Electronic Health Records system has been examined in this study. It proposes a step-by-step graphical query interface to allow semi-skilled users to write queries. Its aim is to decrease user effort and communication ambiguities, and increase user friendliness.

In this paper, we performed six human movement simulation by a commercial software (Poser7). We performed FDTD simulations for body area network propagation with one transmitter and six receivers. Received amplitudes were calculated for every time frame of 1/30 s interval. We also demonstrated a polarization diversity effectiveness for dynamic wearable body area network propagation.

An Energy-efficient Flexible Beacon Scheduling (EFBS) mechanism is proposed to solve the beacon collision problem in cluster-tree healthcare systems. In EFBS, after clustering, BAN Coordinators perform power control. Then they are divided into groups and each group is assigned one contention-free time-slot. The duration of the beacon-only period is flexible. According to the simulation results, EFBS provides better performance than other beacon scheduling approaches.

In this letter, we develop an anycast-based emergency service for healthcare wireless sensor networks. The new service could operate with sensors to detect and activate an alarm system based on predefined conditions that are specific to the patient and the disease. The new service is implemented on 802.15.4 ZigBee which usually has large control overhead and long transmission times. To improve the service performance, our service identifies the closest emergency service provider to alleviate the control overhead and achieve immediate assistance when a patient requests for an emergency service. We also increase the reliability of packet transmission by using a Recovery Point scheme. Our simulations and experiment demonstrate that our scheme is efficient and feasible for healthcare wireless sensor networks.

Body Area Networks (BAN) can provide a wide range of applications including medical support, healthcare monitoring, and consumer electronics with increased convenience or comfort. To harmonize with the strong demands from both medical and healthcare societies, and information and communications technology (ICT) industries, IEEE 802.15.6 task group (TG6) was set up to develop an IEEE wireless standard on BAN. This paper presents a general guidance to TG6. Some pre-works to set up TG6 are reviewed. The objectives, main topics, current status are described in details.

Management of diet and exercise is especially significant in preventing "lifestyle-related diseases" for patients and subclinical cases. This paper introduces a questionnaire survey on diabetic regimens that targets 38 professional users such as physicians and nurses at a diabetic clinic. Based on the results of the questionnaire survey, a design concept for a wearable healthcare support system has been developed to provide patients with timely instruction in accordance with their current context. On the basis of this design concept, we developed a prototype of a wearable healthcare support system called "LifeMinder". "LifeMinder" is composed of a wristwatch-shaped wearable sensor module and a personal digital assistant (PDA). The sensor module measures 3-axis acceleration, pulse rate, galvanic skin reflex (GSR), and skin temperature. The PDA receives this data via BluetoothTM and recognizes the patient's general behavior such as "walking" or "eating". The recognition of these behaviors reduces the patient's mental and physical burden in daily healthcare and assists in support of medical treatment.