(1) Many medical devices that use electricity, such as defibrillators, AEDs, and electrocautery, are used in the medical field. If these devices are not used properly, they can cause electric shocks and burns. In this study, we investigate the electrical hazards associated with improper use of these devices.

(2) It is important to obtain information on patients noninvasively in blood purification therapy. We are developing and improving sensors such as urea sensor using chemiluminescence.

(3) In training clinical engineering technicians, it is necessary to provide various practical training, but it is not possible to use actual patients. Therefore, we are developing educational materials for practical training that can be used in place of living bodies. We are also developing educational tools that can be used in primary life saving seminars for the general public.

We are developing and researching systems and devices that support treatment. Our basic approach is to develop products that operate with mechanical and electrical mechanisms such as robots and manipulators. I am interested in researching systems and devices that read human movements and biological signals in some form and operate or are controlled based on these.

Research on Multimedia system
(1) develop a 3D map presentation system while linking to the database system, and develop applications based on the system
(2) develop a teaching material presentation technology using database ECA rule mechanism

Research on IoT related technologies
(1) devise a location estimation method using radio strength of Bluetooth and evaluate the effectiveness by prototype system
(2) devise a recognition method of human movement status by acceleration sensor data

[Supercomputer × Bigdata × AI] With the experience accumulated through the operation and use of large-scale computer systems, we are conducting research aimed at designing intelligent computational algorithms for bigdata analytics, and creating practical and social solutions from the user’s perspective. Examples of bigdata processing using supercomputer include (i) detection, diagnosis, and prediction of lesions in clinical dentistry, (ii) identification and tracking of individuals in ecological research, and (iii) design of adaptive job scheduling algorithms to reduce waiting time and operational costs. We are developing a wide variety of applications using artificial intelligence, machine learning, evolutionary computation, and combinatorial optimization techniques.

We are developing a system, on which we can operate PC’s application by gesture.
We use the Kinect sensor which can track 25 joints per person every 1/30 second.
In the beginning, we make a system to adapt some gestures with PC’s mouse functions.
For instance, the location of HAND_RIGHT is adapted to mouse cursor, closing right hand is adapted to right-button click, and so on.
We are developing a man-machine interface by using human gesture.

On the basis of the formal language theory, algorithms to solve the language phenomenon have been developed. As applications, my study is developing reputation analysis of the questionnaire, analysis of the words of difficulty degree of difference between the multilingual, and a regional information analysis utilizing short comments (Twitter).

More than ten thousands people died caused fatal accidents in houses every year in Japan. Among them a number of the aged people died in falling down to flat floor even though without any steps. Floor slipperiness is usually represented as static friction coefficient at the slips in toe-off during last half of gait cycle. However, it is said that slips are also occurred at heel down to floor during first half of gait cycle. In order to evaluate floor slipperiness at heel down, it is necessary to measure the dynamic friction coefficient between footwear and floor under incremental vertical load during the heel down.
The wear of UHMWPE(ultrahigh molecular weight polyethylene) used in artificial joint is investigated more than 5 million cycles using joint simulator. The simulator is reproduced vertical load, flexion/extension, abduction/adduction and internal/external rotaion during gate cycle.

Computer games have grown along with technology, incorporating various cutting-edge technologies. Among them, our research focuses on real-time rendering technology that supports interactivity and immersion, and networks that support connections with others. Through research on the mechanisms of getting hooked on games and social media, I would like to work on social contribution through games (education, community revitalization, aesthetics, public relations, etc.) and the role that games play in society.

We are interested in the effects of the artificial environment of imaging systems on humans. We have studied both pros and cons effects. Examples of the good effects are viewers’ sense of power, immersion, and presence, in which the viewers feel as if they are in the space projected by the imaging system. The problem of quantifying these senses is that it is not possible to measure them with a measuring instrument. These characteristics related to human senses and sensitivity can quantified through subjective evaluations and questionnaires. At the same time, biometric measurements are performed to search for measurement items that may serve as biological indices. We are investigating the relationship between subjective evaluations and biological indices.
Conversely, undesirable effects are that a person watching 3D images or a shaky image may experience eye fatigue or a motion sickness. To clarify what can be done to minimize these undesirable effects, we are studying the relationship between the undesirable effects and presented images by subjective evaluations and biometric measurements. In this process, we measure eye movements, autonomic nervous system, and brain activity to quantify the degree of the undesirable effects. Based on these results, we have proposed measures to prevent and mitigate undesirable biological effects.

There are many application fields for the development of rational methods for inferring desired information from observed data. One of the techniques is the alternating least squares method for decomposing a matrix. This method is used in a wide range of fields from analysis of psychological data to image processing. However, the theoretical background of this approach does not seem to be fully clarified. I am conducting research with various interests, centering on research to clarify the theoretical background of the alternating least squares method.

Digital games are both industrial products and a kind of comprehensive work of art, and engineering methods cannot be applied directly to their planning and development. The planning and development of digital games requires an approach that balances creative appeal and industrial quality. We are researching the themes of what kind of production is necessary to achieve this and how the project should be managed.
A game is a kind of system, and a game system is formed by a complex of many systems. We are also researching this system called game mechanics.

The reinforcement learning (RL) learns the behavior through trial-and-error interactions with a dynamic environment. Since RL does not need the precise infromation on the desired output or desired action, it is considered to be applicable to many types of control problems. In order to develop the effective learning algorithm, we use the multiple actions as an output, and INGnet to process the sensor inputs. This learning system shows faster learning (left panel).

The signal transduction pathway (STP) is a sytem of intracellular reactions to govern the control of homeostasis. We are developing the computer model of immune system using STP to investigate the molecular mechanism of the emergence of diseases. Recently, we are developing the computer model to mimic the emergence of rheumatoid arthritis (right panel).

Even if you think you understand it, if you look at it from a different perspective, you may find some unclear points.
We are researching the construction of an easy-to-understand learning system that makes difficult-to-understand teaching materials easier to handle, encourages confirmation from multiple perspectives, and is easy to understand.
Based on conventional CAI and e-learning methods, we aim to build and release learning support systems such as algorithms and programming.
We are also researching a system that presents answers to simple puzzles in order to cultivate logical thinking.

１. It is important to develop the orthosis which improves the Quality of Life (QOL) and maintains health conditions. We focus on an electric wheelchair which is a welfare apparatus for mobility support of a physically handicapped person. Therefore, the estimated method can be obtained unknown parameters of the driving environment.

２. It is important to develop the orthosis which improves the Quality of Life and maintains health conditions. As one of treatment methods done to lumbago, the waist fixation method is performed. A waist active orthosis implement with pneumatic flexible actuators developed.

３. We develop 3D quasi-passive walker with flat feet.
And an antagonistic pneumatic system is used as joint actuators. 3D quasi-passive walking in the level ground can realize by the simple swinging input of the frontal direction.

By taking advantage of pneumatic actuators that have features of higher force / mass ratio, lightweight and flexibility based on compressibility, we have developed low-cost wearable actuators and low-cost home-based rehabilitation devices that can be disposable after using it. Specifically, we developed a flexible pneumatic cylinder that can be used even if the cylinder bends. A spherical actuator that consists of two flexible pneumatic cylinders have been also developed as an application for portable rehabilitation device. We have also developed a low-cost wearable servo valve using buckled tubes, an artificial muscle with built-in displacement sensor and a pipe inspection robot using extension type flexible pneumatic actuators. In addition, aiming at globalization for students, all graduate students in our laboratory have presentation in international conferences.

(Numerical analysis)In scientific computing most computational time is spent in solving systems of linear equations. It is necessary to study an algorithm for solving the large linear system fast and stably. We have proposed several new iterative algorithms and preconditioning techniques.

Vertical axis is the number of iterations, horizontal axis is the index of residual vector, and we display the magnitude of the residual vector in the color. When this becomes black beyond the red, the error becomes large and when this becomes white beyond the yellow, the error becomes small.
(ICT)Distance learning is a way of learning remotely without being in regular face-to-face contact with a teacher in the classroom. We study the support system for the programming lecture in distance learning.

In recent years, computer CPUs have become more high-performance, and the number of CPU cores has been increasing year by year to 8, 12, 16, and so on.
As the CPU increases the number of cores, the software needs to be parallelized in order to obtain high performance and high throughput
dataflow.
This situation applies not only to high-performance computers but also to low-priced, limited-performance computers such as smartphones,singleboard computers or IoT devices.
In our laboratory, we are conducting research on the topic of that parallelized microservices using OS virtualization technology on computers attached to small sensor devices such as IoT and WoT.
On network devices, the scope of research is further, and the topic is packet and path control corresponding to traffic increase due to parallelization.

Researchs are being conducted on care and welfare robot systems using soft pneumatic actuators and embedded controllers that can be used in ordinary homes. One of my research is the upper body support device for caregivers inclouding the part of arm will improve human affinity and do not interfere with caregiving movements. Because of an inexpensive, lightweight, and waterproof embedded controller, the support device can be used in small and humid spaces. The other one of my research is about the the walking training device for visually impaired children that is to reduce the burden on walking trainers, who are facing a shortage of manpower, and to reduce the physical and mental burden on the guardians. In addition, in my lab, we also develop the mattress that uses a flat soft actuator to change the position of contact between the body and the actuator by varying the applied pressure over time. Because when caring for the elderly and severely disabled, changing positions to prevent bedsores for bedridden patients which increases the caregiver’s restraining time and increases the mental burden.

The theme of our laboratory is to measure biological information obtained from humans and animals. Measured biological information is used in the development of various sensors, equipment, and software.

Statistics with Computer, R, Python, etc.