With the number of dementia patients worldwide projected to approach 115 million by 2050, the development of simple early diagnosis tools to prepare for the rapid increase in the number of dementia patients is expected. Therefore, we are developing a dementia screening tool using a character input type BCI (Brain-Computer Interface). Our research group has already observed cognitive decline from an increase in Spelling-Error Distance Value (SEDV), but we did not find a statistically significant difference between the group with cognitively unimpaired participants (CU group) and the group with mild cognitive impairment (MCI group), which is necessary for early diagnosis. In this study, we focused on Tsallis entropy and coherence to identify EEG features that explain CU group, MCI group, and Alzheimer's disease patients (AD group). The data from 55 participants were analyzed. The MMSE, a measure of neuropsychological testing, was 28.20, 27.38, and 22.04 for the CU, MCI, and AD groups. The results showed that Tsallis entropy decreased and coherence increased with cognitive decline. Significant differences (p<0.1 and p<0.05) were confirmed in multiple comparisons between the CU and MCI groups for both indices, indicating that progression from CU to MCI was predictable.

View full abstract

There are growing expectations that wearable devices will provide feedback on the health status of older adults, which can be used to raise their awareness of voluntary health maintenance and appropriate intervention by their supporters. On the other hand, older adults have many difficulties in the use of information devices, and at the same time, it is difficult to find out their own ways to use ICT devices or values of their usage, for older adults. In order to investigate those possibilities, a three-week continuous usability testing experiment was conducted for 8 healthy older adults using a wristband-type health smartwatch (Fitbit Sense). As a result, all participants were able to use the device continuously for three weeks, at least in the circumstances where we the testers set up. Even though their activity indexes were not changed significantly, there were some focused interests by older adults in their activities and/or daily reports; interests to the sleep evaluations, motivations to physical activities, and awareness of time and executing activities. Importance to design for being able to set-up their using circumstances by older adults themselves, and ways to sense and keep users’ interests or their values to use will be the keys to wearable information devices can be tools for older adults.

View full abstract

Individuals with cognitive impairments usually have difficulty in perform ing sequential tasks and instrumental activities of daily life, and need external help with rehabilitation. While traditional therapies, e.g. on-desk assembly work and puzzles, are effective in improving the performance of each specific ability, it still remains unclear if the patient is sufficiently recovered for practical scenarios. This research is aimed to explore the efficacy of cognitive tasks using immersive virtual reality (VR) , especially focusing on 10 different sub-functions regarding executive functions and attention. We implemented a system of an indoor post office scene, with various in-situ tasks that require the user to correctly evaluate the situation to make proper decisions. We conducted a user study over 17 participants. The results indicate that the cognitive task using immersive VR is able to evaluation the cognitive functions comprehensively.

View full abstract

Head-up display (HUD) is an automotive driver assistance technology that displays images on the windshield to present important information, whenever necessary. Drivers process this visual information by examining the HUD images captured from the road environment. Therefore understanding the input process is important for designing a user-friendly HUD. In this study, we focused on the relationship between the HUD’s video projection method and video display distance. Eye tracking method was used to examine the impact of different configurations on the process. Consequently, we observed that with an increase in the video display distance, the processing time becomes shorter in the virtual image display contrary to the real image display, where we found no such change. Therefore, there are two important parameters for input processing: display distance and ease of spatial orientation.

View full abstract

In deep learning, large and accurate supervised datasets are important. Datasets are created by organizing source images and annotating them with information according to the task to be trained using machine learning. However, if a dataset is not available, the creation of a supervised dataset is required in addition to the usual work. Software to assist annotation has been proposed, but it requires work in addition to the usual tasks. We propose a deep learning dataset generation cycle in which the annotation work is included in the image measurement work. In this paper, a deep learning dataset generation cycle was added to the image measurement support tool TouchDeMeasure and developed. The proposed system was evaluated by comparing the annotation time and annotation results of conventional and proposed methods for a sampling survey of clams. The results show that the proposed method reduces the annotation time to approximately 2 to 3 seconds, while the previous method takes approximately 10 to 19 seconds per annotation. Using the dataset obtained by the proposed method, we constructed and evaluated a learning model using Mask R-CNN and obtained 99.2% accuracy, demonstrating the effectiveness of the proposed method.

View full abstract

A number of studies have suggested a relationship between visual attention and microsaccades, a type of fixational eye movement. Visual attention can be classified as Top down attention and Bottom up attention. However, the effect of these types of attention on microsaccades has not been clarified. Further, the inclusion of rapid saccades in research on eye movement for practical applications warrants evaluation. In this paper, we investigated the effect of Top down attention and Bottom up attention on microsaccades extracted from eye movement data which includes saccades. Eye movement, pupil diameter and response time were measured under five controlled experimental conditions requiring different processing loads. The image features, pupil diameters and response times indicated that the experimental conditions were properly configured. Statistical analysis indicated that microsaccades showed different behaviors under each of the five experimental conditions. This result suggested that visual attention and processing loads affect behaviors of microsaccade, and can be used to estimate the required processing load by visual attention.

View full abstract

We aim to build a support environment for software engineers to improve usability for various stakeholders of information systems and created its design guidelines to incorporate the concept of human centered design into system development process. Based on these guidelines, the method for clarifying user needs concerning usability, which is the core of the support environment, was developed. The support environment incorporating this method was evaluated by three main roles of development teams: project managers, usability experts, and software engineers. Through the evaluations, the usefulness in quality and cost from the perspective of project management was verified, and positive effect on work efficiency and quality compared to manual work was verified when it was used by usability experts. It was found that software engineers may be able to extract "HI requirements (design goals concerning usability)" which have large impact on design man-hours by using the support environment. Furthermore, two requirements for the environment to be more widely used in development teams were identified: expanding its guidance for navigating work to clarify user needs and improving its information expressions such as confirmation items. To the future, we will focus to user requirements includes other stakeholders (e.g. indirect user or non user).

View full abstract