Data science is the study of extracting knowledge and making inferences from data. Various data science techniques have been developed to manage, process, and use big/heterogeneous data. The most successful technique is deep learning using neural networks. This paper first describes the current state of machine learning. It then presents two applications of machine learning are presented. The first is an application of machine learning in molecular biology, and the second is an application of machine learning in biological signal processing. The molecular biology application will mainly explain DNA methylation analysis will be explained. The biological signal processing application will mainly explain the classification and quantification of tremors.

Studies on electroencephalography-based brain computer interface (EEG-based BCI) has been widely studied since around 2000. In the studies, three methods, event-related desynchronization/synchronization (ERD/ERS) during a motor imagery task, a P300 component of event-related potential, and steady state visual evoked potential (SSVEP) has been studied as standard methods. Since around 2013, with the acceleration of research in the United States and other countries, there has been a wide variety of studies that is not limited to the standard methods, such as research and development toward practical application and studies on extracting new brain activity signals. Some of the research results have been developed as commercial products. In this paper, after reviewing the studies on EEG-based BCI, topics in the research and development toward practical application such as new EEG systems and artifact rejection methods are introduced, topics in the studies on extracting new brain activity signals such as applying deep learning, emotion estimation and visual image reconstruction are also introduced, and topics in combination of BCI and information and communications technology and combination of BCI and XR technology are introduced. Finally, prospects of BCI technology are discussed.

This study aimed to test the efficacy of a mint-flavored mouthwash when it is used at bedtime for subsequent sleep in a field study setting, using a wristwatch-type heart rate (HR) monitoring devise. Using a within-subject experimental design, twenty healthy adults used three types of mouthwash having a different intensity of mint sensation (or water as a control) per one experiment night just before going to bed at their home environment, in a counterbalanced order. As for results, irrespective of the intensity of mint-flavored mouthwashes, the subjective score for “fatigue” and “refreshing” before sleep was decreased and increased, respectively. In addition, although a significant inhibition of HR decline in the sleep initiation period (0-30 minutes after bedtime) was observed with the use of mint-flavored mouthwashes compared with the control, the trend reversed subsequently as HR trended lower than for the control during 2-6 hour after the bedtime. These findings suggest that the using mint-flavored mouthwashes at bedtime induces a positive mood before sleep and may improve physiological sleep. Moreover, this study illustrated the possible application and significant advantage of the use of wristwatch-type heart rate monitoring devise in a context of field study.

In recent years, various studies on anthropometry using video images have been actively conducted. Of those applications, the measurement of eye movements is expected in fields such as medical care, and human interface. The eye movements include horizontal and vertical translations due to the movement of the gaze and torsional eye movements. This paper proposes a method for accurately measuring the torsional eye movement, which is generally considered difficult to measure with high accuracy. We have improved the Monte Carlo filter, which was traditionally used for object tracking in translational motion, so that it can also measure rotational motion. Evaluation experiments indicated that the measurement error using the proposed method was in a range between 0.07±0.49° to -0.03±0.54°, while the measurement accuracy of the conventional matching method was 0.28±0.81°. In addition, the results suggest that the proposed method can measure even a blurred iris image with a relatively small error.

In this study, as a basic investigation for EEG-based visual image reconstruction, we investigated whether EEG signal features reflect shapes and colors of simple visual images which subjects viewed and whether the features can be discriminated. First, we investigated how the shapes and the colors are reflected in event-related potentials (ERP), the event-related spectrum perturbations (ERSP), and the inter-trial phase synchronization (ITC). The results showed statistically significant differences in ERP among the colors and ERP, ERSP and ITC among the shapes depending on time periods, frequency bands and electrodes. Second, based on the results, we explored learnable input data sets. Then, learnability for discriminating the shapes were shown in EEG waveforms on 100ms time periods in single trials at all channels and phase of time frequency analysis on the limited time-frequency domain and electrodes. Finally, we investigated whether two discriminators using LSTM and CNN discriminate the shapes from the learnable data for each subject. Then, it was found that accuracies of discrimination of the shapes were over a chance level with all the learnable data sets, subjects, and discriminators. We concluded that the distinct shapes can be discriminated from EEG signals by exploring appropriate features of input signals for discriminators.

The direction of body sway induced by neck dorsal muscle stimulation can provide useful insights into roles of sensory information in human postural control. Previous studies suggest the existence of a common brain mechanism to determine the body sway direction in response to the muscle stimulation. However, in those studies, direct evidence has not been obtained because of the limitation of experimental techniques. In this paper, experimental methods were proposed to simultaneously measure attention and body sway direction, aiming at obtaining more direct evidence. Line-motion illusion was employed to estimate visual-attention while binaural separation hearing test was used to estimate auditory-attention. The experimental results indicated that the body sway directions were significantly different in different condition of attention. These results demonstrated that the direction of the attention plays an important role in determining the direction of body sway.

Skeletal muscle is a set of motor units (MU). Muscle contractile activity is carried out by regulating the firing frequency of MU and the types of muscle fibers to be mobilized. Multi-channel surface electromyogram (sEMG) contains many conducting waves that represent a single motor unit action potentials (MUAP). In previous study, we proposed a method to extract conducting waves quantitatively and automatically. This method is useful for elucidating mobilized MUs and can be applied to kinematic analysis and diagnosis of muscle diseases. In multi-channel sEMG measurement, rows of electrodes need to be applied along the direction of the muscle fibers. However, the electrodes are difficult to set because the direction of the muscle fibers cannot be visually confirmed on the skin. This study investigated a method for estimating muscle fiber direction by analyzing conducting waves from two-dimensional multichannel surface electromyograms using grid-shaped surface electrodes. By examining the number of conducting waves acquired in each direction of the electrode row, it was suggested that the direction of muscle fiber may be estimated from the direction with the largest number of propagating wave acquisitions. It was also found that the method of acquiring differential potential signals has a significant impact on the analysis of propagation direction.

We developed an analyzing system for tremor using the three-axis accelerometer built into the Wii Remote in 2008. We propose a theory of estimating the tremor amplitude and mounted it in our system. The aim of the study was to study of errors in tremor amplitude estimates. To test our system using a small accelerometer (WAA-006), mechanical oscillations generated by an oscillation generator were sent to a small accelerometer that was attached to the device. The oscillating amplitude was set at 0.5, 1.0, 1.5, and 2.0 cm, and the oscillating frequency was varied from 1 Hz to 15 Hz in steps of 1 Hz. The relative error of the amplitude estimates calculated from the accelerations measured by the WAA-006 was 103.9 % at 3 Hz and 1 cm. Since this experiment covered the measurement range of the accelerometer, we were able to present the limits of amplitude measurement and analysis by the accelerometer.

It is significant to evaluate the recruitment of fast twitch fibers quantitively in means of preventing sarcopenia. In a previous study, m-ch method was proposed to observe propagation wave which originates from action potential of each muscle fiber. In this study, relation between strength of muscle contraction and propagation wave had been examined in purpose of evaluating recruitment of fast twitch fibers during training. The result suggests that most propagation waves acquired band's velocity has increased by increase of muscle contraction strength. There is a hope for evaluating recruitment of fast twitch fibers during training.

This study examined the differences in functional brain network over time between different anxiety states and evaluated their usefulness in neural networks (NN). Seventeen young adults with high-anxiety and 13 young adults with low-anxiety were examined. The subjects were given three stimulations: resting, pleasant, and unpleasant stimuli, and Electroencephalogram (EEG) was measured immediately after the stimuli. EEG was analyzed for the alpha band using coherence analysis and graph theory. We evaluated the classification accuracy of anxiety states by NN and recurrent neural networks (RNN). The results showed the information processing process and structure of the brain functional network to emotional stimuli differed over time depending on the anxiety state. The time series data of coherence and graph theoretical indicator by EEG would be considered to be useful for discriminating anxiety states using RNN.

Conventional recommendation systems such as for books or movies are primarily based on purchase and browsing history, which do not reflect user evaluation. This study focuses on analyzing invisible and feeble physiological tremors to reflect user evaluation. While physiological tremors are generally measured by fixing joints and wearing a light accelerometer, considering implementation, we measured them without fixing joints. This study aims to examine whether it is possible to acquire acceleration data including physiological tremors with no limitation on the posture and extract features reflecting the user interest. In the experiment, subjects read comic books on a smartphone and the acceleration data were collected using the 3D accelerometer of a smartphone. Assuming the actual use environment, subjects read comic books in both sitting and standing with no instruction on their posture. As a result of performing fast Fourier transforms, it was suggested that the acceleration data included physiological tremors and could estimate user interest while reading.

To improve the proper diagnosis rate beyond that of sweep-frequency tympanometry for auditory ossicle examination, a new method adopting a stepwise-variable-frequency and a wide-band frequency of probe tones is proposed to be used in place of the conventional method. It is shown that there are similar trends in the measured properties when the proposed method is applied to an artificial model with the external auditory meatus, tympanum, and auditory ossicles and to a subject. There are differences in examinational indexes of the auditory ossicles when the method is applied to the artificial model with and without a push-spring for the disarticulation of auditory ossicles. The results indicate that the proposed method is promising.

Many recommendation methods for cooking recipes are proposed in previous research. There are many studies on maximizing evaluation score on healthcare, but few studies on maximization under the limits on the cost of purchasing additional ingredients and food loss. In this research new recommendation and purchase methods are proposed. Dynamic programming is applied in the proposed methods. The basic proposed method maximizes the evaluation score on healthcare under the limits on the cost of purchasing additional ingredients and food loss. It is possible that there are multiple optimal solutions. To address this issue, two improvements of the basic proposed method are implemented. The two improved proposed methods minimize the additional purchase cost and the food loss in addition to the maximization of the healthcare score when there are multiple optimal solutions. The minimization of the additional purchase cost takes priority over the minimization of the food loss in the first improved proposed method. The minimization of the food loss takes priority over the minimization of the additional purchase cost in the second improved proposed method. The effectiveness of the proposed methods is shown by some computational examples.

We found that some signs of Mild Cognitive Impairment (MCI) might be presented in a structure of a sentence and a relation between sentences talked by a man, and develop a neural network model which has an analogy with the hierarchical structure of speakers, tpoics, sentences and words in Japanese. We build our model based on 2-layered bi-directional LSTM, corresponding to words-sentences and sentences-topics hierarchy. As a layer corresponding to speakers, we use a linear classifier with self-attention. The test result shows a largely improved AUC, compared with our another test by using the normal 2-layered bi-directional LSTM with TBPTT. The result also indicates that there are some characteristic patterns in a talk by an elderly person with MCI. We classify the character vectors of topics generated from our model through learning into clusters whose number is 1/10 of the number of persons in our data. Since these clusters have almost less than 10% or more than 90% rate of positive share, we conclude that we can develop a screening method based on a talk in Japanese by an elderly person in the near future.

In hemodialysis, normal venipuncture cannot deliver large amounts of blood, so about 90% of patients make an arteriovenous anastomosis called an internal shunt. In addition, since dialysis is performed multiple times a week, frequent puncture and pressure hemostasis are required. As a result, blood vessels are stressed, resulting in stenosis and occlusion. In order to detect and treat such shunt troubles at an early stage, daily shunt management is important. Laser speckle contrast analysis is of interest in the field of visualization of the hemodynamics of living tissues. In this study, the relationship between the flow rate and the contrast value was investigated by a technique for quantifying the amount of change in the dynamic speckle pattern.

It is important for disease control to grasp the amount of walking in stroke hemiplegic patients. However, it is often difficult for hemiplegic patients to measure their steps accurately with commercial pedometers. Our aim is to develop an acceleration measurement system for hemiplegic patients. In this study, we investigated the feasibility of an acceleration measurement system using a smartwatch (Fitbit Versa., Fitbit, Inc.) as an acceleration sensor and a smartphone as a data recorder.

Recently, a high-efficiency solution of combinational optimization problem using Ising machines has been studied for many labs. We have been proposed an Ising machine using parametric resonant circuit with capacitor and inductor. We took a simulation of the Ising model using these circuits, which have been shown in the reproducing of the feature physical phenomenon in the ferromagnetic and the antiferromagnetic Ising models. Furthermore, we calculated the ground-state-search of the Ising model with 100-spins circuits, which succeeded in solving the ground-state. From these results, we suggested that there was able to imitate the Ising Machine using the parametric resonance circuits.

In recent years, the lowing supply voltage and the precising accuracy are required for ΔΣA/D converters. The swing-suppression circuit is proposed to reduce the supply voltage to nearly 1 V. However, this circuit cannot realize the high order shaping characteristics over the second order, and it is difficult to improve the accuracy. In this report, we propose the accuracy improvement method using MASH structure. Using 2-1 MASH structure and the 2^nd order swing-suppression A/D converter. To realize the third order shaping characteristic by the proposed method, the SNR of 86 dB are realized.

Spatial attention disorder, also known as unilateral spatial neglect (USN), and non-spatial attention disorder often co-occur after stroke. In previous studies, the development of an electronic USN test system has made it possible to record detailed test data that are impossible with paper-based testing. However, the method of providing information on the test data has not been investigated, and the USN test system is exclusively used for research purposes. Therefore, we developed a system to provide information on the models of attention disorders in different formats, such as numerical, graphical, and superimposed images. The developed information provision system can provide information according to the characteristics of the users. Evaluation experiment was confirmed that the system is effective to change the information provision format according to the subject.

An extended-gate field effect transistor (EGFET) based creatinine sensor using an enzyme-containing silk-fibroin membrane successfully fabricated. In this experiment, we used creatinine deiminase as a detection enzyme for serum creatinine, which is known as a marker of chronic kidney disease. The fabricated biosensor showed excellent potentiometric response to creatinine over a wide range from 0.007 - 0.4 mg/mL in phosphate buffer solution, which can enable to measure creatinine concentration in blood plasma.

In this paper, a new data-driven control is proposed to realize desired closed-loop responses. Proposed methods can tune the feedback controller by estimating desired control signals. The evaluation function can be minimized by least square method because the inverse model of the feedback controller is not required. The validity of the proposed method is verified via numerical simulation and experiments.

Meniere’s disease, a type of inner ear disease, is thought to be caused by ischemic lesions in the inner ear. On the other hand, Meniere’s disease is often associated with sleep apnea syndrome, and the relationship between the two has been pointed out. In recent years, many patients with Meniere’s disease have shown improvement in their symptoms after discontinuation or suppression of medication and sleep therapy. In this study, we hypothesized that the Electroencephalogram (EEG) during sleep in patients with Meniere’s disease has a characteristic pattern that is not seen in normal subjects. The EEGs of normal subjects and patients with Meniere’s disease were converted to lower dimensions using a variational auto-encoder (VAE), and the existence of characteristic differences was verified. Sub-sequence was extracted from the EEGs of 20 subjects, which was input to a variational autoencoder and was converted to lower dimensions. The machine learning was conducted for each channel. Latent variables obtained from the VAE were classified using Support Vector Machine (SVM). The results showed that the electrodes located at the back of the head had a higher correct response rate and F value.

Legacy enterprise information systems are required to convert their implementation languages into modern languages to enhance their maintainability. However, in a planning phase, it is difficult for engineers to identify units of programs to be converted appropriately and quickly. In this paper, we propose a support method that can effectively identify the units of programs for language conversion and test data acquisition points by relating application logs to software structure. We implemented this method by using graph database, and through the application of this method to real enterprise information system, we confirmed that we can effectively identify units of program conversion and test data acquisition points.

CVT-MAP-Elites is a method of evolutionary computation that maintains diversity and searches for solutions in the feature space by maintaining the optimal solution in the segmented feature space, but there is a problem of ensuring a sufficient number of searches for local search. In this paper, we propose a local search method by transcribing partial solutions of elite individuals and report the results of applying it to multi-tasks optimization problems.