Faculty Profiles - YAMAZAKI Tatsuya

Papers - YAMAZAKI Tatsuya

Division display 1 - 20 of about 103 ／ All the affair displays >>

CT Image Diagnostic Support System for Predicting EGFR Gene Mutations in Lung Cancer.

Cher Yen Tan, Ryotaro Akagawa, Tatsuya Yamazaki, Motohiko Yamazaki

Journal of Information Processing 32 ( MBL-108 ) 533 - 542 2024

Publishing type：Research paper (scientific journal)

Lung cancer is the most common type of cancer and is the leading cause of cancer-related deaths in Japan. Regarding lung cancer diagnosis, a pivotal aspect in lung cancer treatment is tumor detection and selection of appropriate cancer treatment. Computer Tomography (CT) images are usually used to detect tumors. After tumor detection, identifying mutations in the Epidermal Growth Factor Receptor (EGFR) gene is essential for cancer treatment selection. The EGFR gene is a key factor associated with cancer cell proliferation its mutation appears both inside and around the tumors. This paper proposes a lung cancer diagnostic system designed to streamline the process from tumor detection to EGFR gene mutation identification. The proposed system consists of three modules: an input interface module, an automated lung tumor segmentation module, and an EGFR mutation prediction module. The system is characterized in that all modules are consistently based on image processing. Consequently, the proposed system enables users to acquire the diagnosis results of tumor detection as well as EGFR gene mutation prediction by just providing the input CT image data and the patients’ clinical features. Our experimental results confirm that the system achieves performance levels comparable to existing research, both in terms of lung tumor segmentation accuracy and the precision of EGFR mutation predictions.

DOI： 10.2197/ipsjjip.32.533

Machine learning algorithms mimicking specialists decision making on initial treatment for people with type 2 diabetes mellitus in Japan diabetes data management study (JDDM76). International journal

Jenny Elizabeth Price, Kazuya Fujihara, Satoru Kodama, Katsuya Yamazaki, Hiroshi Maegawa, Tatsuya Yamazaki, Hirohito Sone

Diabetes & metabolic syndrome 18 ( 11-12 ) 103168 - 103168 2024

Language：English Publishing type：Research paper (scientific journal)

OBJECTIVE: To evaluate whether typical machine learning models that mimic specialists' care can successfully reproduce information, not only on whether to prescribe medications but also which hypoglycemic agents to prescribe as initial treatment for type 2 diabetes. RESEARCH DESIGN AND METHODS: A medical records database containing prescriptions for medications for 16,005 patients who visited a diabetologist's office for the first time was utilized to train five typical machine learning models as well-as a model used for logistic analysis. Prescribed were no medications (diet and exercise therapy), insulin, biguanides (BG), sulfonylureas (SU), dipeptidyl peptidase-4 inhibitors (DPP-4I), alpha-glucosidase inhibitors (α-GI) or glinides. Models were compared based on the F1 score and ROC/AUC scores. RESULTS: XGBoost, which splits decision-making into three sections, was the top performing model (42 % accuracy) among five models and conventional logistic regression (35 % accuracy). The second highest scoring model was Support Vector Machines, which had an accuracy of 40 %. When using XGBoost to compare decisions on no medication needed vs. needing medication the AUC was 0.96. Insulin vs. oral medications had an AUC of 0.78. With all remaining oral medications removed, the AUC was 0.76. CONCLUSIONS: Among the five models investigated, XGBoost outperformed the other machine learning models examined as well as the traditional logistic model, suggesting that its accuracy had the potential to assist non-specialists in decision-making regarding treatment of patients with type 2 diabetes in the future.

DOI： 10.1016/j.dsx.2024.103168

Development of Automatic Visual Anomaly Detection System for Data Centers

Misheel Enkhbaatar, Tatsuya Yamazaki

Sensors and Materials 36 ( 6 ) 2539 - 2556 2024

Publishing type：Research paper (scientific journal)

In this paper, we present a practical automated visual monitoring system designed to enhance the efficiency of visual inspection in data centers. Visual inspection is a manual process of detecting failures in electronic devices based on light-emitting diode (LED) lighting. The objective of data center monitoring is to implement real-time failure detection to prevent any service disruptions or loss of user data. To improve the reliability of data centers, we propose a monitoring system that automatically detects anomalies in electronic devices. The system integrates a digital camera and a novel algorithm that is tailored to distinguish normal LED lighting patterns from abnormal patterns. Experimental data were collected in an actual data center room and the system was evaluated with experiments involving LED region segmentation and anomaly detection. For the LED segmentation task, we propose a K-means-based method that outperformed a previous method based on background subtraction by 8%. For anomaly detection, recorded videos covering continuous monitoring of approximately 17 h were used. The proposed method successfully detected all five true anomalies in the video data. The results of another experiment for anomaly detection demonstrate that prolonged video recording for collecting patterns of LED lighting can positively contribute to a better understanding of normal patterns and can effectively be used to ensure the detection of device anomalies.

DOI： 10.18494/SAM5012

Fundamental Sentiment Analysis by Natural Language Processing and Machine Learning for Email Classification.

Adewale Adesoji Emmanuel, Tatsuya Yamazaki

Proceedings of the 2023 5th Asia Pacific Information Technology Conference(APIT) 2023 103 - 105 2023

Publisher：ACM

Due to its ease of use, speed, adaptability, and ability to keep a complete record of correspondence, email is a commonly used and trusted communication medium. The vulnerability of these emails to cyberattacks has increased. This study utilized the hybrid-based sentiment analysis approach for email fraud detection. The lexicon-based using the Word2Vec feature and Machine Learning (ML) classification approach using both the Bag of Words and TF-IDF feature extraction techniques are adopted to optimize accuracy. The ML classifiers adopted are Random Forest, k-nearest neighbors (k-NN), Support Vector Machine, and Logistic Regression. The predictive and classification comparison of the ML models suggests that Random Forest slightly outperformed the other models with 87% accuracy and 78% F1-score and k-NN performed worst.

DOI： 10.1145/3588155.3588171

Other Link： https://dblp.uni-trier.de/db/conf/apit/apit2023.html#AdewaleY23

An Anomaly Detection Method for Data Center Devices

Misheel Enkhbaatar, Tatsuya Yamazaki

2023 6th World Symposium on Communication Engineering, WSCE 2023 28 - 33 2023

Publishing type：Research paper (international conference proceedings)

Data center demand is growing significantly due to the widespread use of remote working by businesses and cloud-based services. As a demand grows, the number of infrastructure devices, such as servers and storage, increase. An interruption or the failure of a device is costly for the data centers. To prevent thousands of electric devices from unpredictable failures, the data center must implement a speedy and accurate monitoring method. A visual monitoring method is considered reliable for detecting failures in the electric devices, but the method is very slow to monitor numerous devices due to fully manual work and a pattern of failures is not explicitly defined. Therefore, automation of the visual monitoring method is required for the data centers. This paper discusses an anomaly detection method from the video recording of the devices and development of a monitoring system. In an experimental environment of Niigata University, the proposed method performed 100% (anomaly: 4/4, regular: 225/225) accuracy for detecting anomalies from the video frames.

DOI： 10.1109/WSCE59557.2023.10365893

Information Security in Social Media Sites: Sentiment Analysis of Email.

Adewale Adesoji Emmanuel, Tatsuya Yamazaki

18th IEEE International Conference on Computer Science and Information Technologies(CSIT) 1 - 5 2023

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

In today's digital era, email stands as a fundamental tool for communication due to its convenience, rapidity, flexibility, and comprehensive archival capabilities. However, the escalating risk of cyber threats targeting emails has introduced new challenges. This research introduces an innovative approach to identify fraudulent emails through a hybrid-based approach using sentiment analysis. The proposed method unites the lexicon-based technique using Word2Vec features with a Machine Learning (ML) classification approach. Two prominent feature extraction methods, Bag of Words and TF-IDF, are integrated to enhance the precision of the analysis. The study employed well-established ML classifiers including Random Forest, KNN, Decision Tree, SVM, Multilayer Perception, and logistic model to analyze the emails. The result of the analysis was impressive across the selected models, with Support Vector Machine and Multilayer Perception slightly surpassing others with an impressive 99% accuracy, F1-score, and Recall rate respectively. This emphasizes the effectiveness of this model in recognizing fraudulent emails.

DOI： 10.1109/CSIT61576.2023.10324038

Other Link： https://dblp.uni-trier.de/db/conf/csit/csit2023.html#EmmanuelY23

Traveler Data Collection and Analysis to Extract Travel Behavior

Tatsuya Yamazaki

International Conference on Electrical, Computer and Energy Technologies, ICECET 2023 2023

Publishing type：Research paper (international conference proceedings)

To enhance the Value of Travel Time (VTT) is important for the tourism-related or transport infrastructure industries. Evaluation of VTT needs a simple way to collect vast and diverse traveler log data. In this paper, an experiment collecting such huge traveler data set by using QR codes and smartphones. Moreover, the static and dynamic data analyzing methods of the traveler log data are introduced. Finally, a typical traveler behavior is extracted from the data.

DOI： 10.1109/ICECET58911.2023.10389556

Quality Evaluation of Kohaku Koi (Cyprinus rubrofuscus) Using Image Analysis

Mikhail A. Domasevich, Hideo Hasegawa, Tatsuya Yamazaki

Fishes 7 ( 4 ) 2022.8

Publishing type：Research paper (scientific journal)

Quality evaluation of koi (Cyprinus rubrofuscus) is essential to the koi industry. However, the community has a lack of knowledge about the quality evaluation of koi and about features of koi which are valuable. During the “All Japan Nishikigoi Show” competition, a quality evaluation of koi was conducted by authoritative experts. In the present study, photos of participants of this competition were used for analysis to discover features affecting the quality of the Kohaku variety of koi. The HSVA color model was used to define and extract from image red and pale red coverage rates in Kohaku body coloration. Body aspect ratio, body proportions, red coverage rate, and pale red coverage rate were extracted using tools written in Python programming language. Analyses of the extracted data revealed the significant impact body aspect ratio (p < 0.001) and pale red coverage rate (p < 0.001) on Kohaku quality. A low body aspect ratio has a positive impact on Kohaku quality, while a high pale red coverage rate has a very strong negative impact on Kohaku quality. This study and its results can be an important foundation for further research about koi quality.

DOI： 10.3390/fishes7040158

Performance Evaluation of Device LED Extraction for Data Center Infrastructure Monitoring.

Misheel Enkhbaatar, Tatsuya Yamazaki

11th IEEE Global Conference on Consumer Electronics(GCCE) 613 - 614 2022

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

Availability is the core metric of a data center operation. To be available at any time, data centers must use a real-time and precise failure detection method. However, current methods are not enough to bring data centers to their goal. One of the practical device failure detection methods is visual monitoring. Human inspectors detect failures by checking LEDs of infrastructure devices. This manual method requires a lot of effort as well as it is not suitable for real-time monitoring for big scale data centers. The goal of this study is to achieve real-time and high accuracy failure monitoring for the data center. This paper presents performance evaluation of an automatic LED extraction system which is a part of work to achieve an automatic data center monitoring. The system performed the LED detection with an accuracy of 83.7%.

DOI： 10.1109/GCCE56475.2022.10014401

Other Link： https://dblp.uni-trier.de/db/conf/gcce/gcce2022.html#EnkhbaatarY22

XGBoost: Interpretable Machine Learning Approach in Medicine

Jenny Price, Tatsuya Yamazaki, Kazuya Fujihara, Hirohito Sone

WSCE 2022 - 2022 5th World Symposium on Communication Engineering 109 - 113 2022

Publishing type：Research paper (international conference proceedings)

This research tries to establish a communication bridge between the decision making of a practicing physician and a machine learning model. Machine learning has long been used to assist in diagnosing patients through deep learning and medical image analysis. However, in medical science, it is possible to have multiple right answers when treating patients. This leaves the practitioners room to exercise their judgement on what would be best for the patient. The machine learning model can be a communication tool to assist the practitioners in reach the final decision. In this paper, the machine learning model used for this experiment is XGBoost, a gradient boosted decision tree. Through the experiment with the real patient data, we show how decision tree constructions are easily readable for medical determination.

DOI： 10.1109/WSCE56210.2022.9916029

A Vision-Based Auricular Diagnosis Assistance System with Deep Learning.

Meng-Lin Chiang, Ling Hou, Chiung-Yao Fang, Tatsuya Yamazaki

4th IEEE Global Conference on Life Sciences and Technologies(LifeTech) 475 - 479 2022

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

In the area of auricular medicine research, researchers believe that human diseases are reflected in corresponding locations on the external ears, known as the positive reactions of auricular points. If some auricular points on a person's external ear exhibit abnormal positive reactions, then the person is being attacked by a corresponding disease. This paper proposes a vision-based auricular diagnosis assistance system to help doctors detect diseases by discovering the visual positive reactions of auricular points on patients' external ears. The proposed vision-based auricular diagnosis assistance system includes visual positive reaction detection and disease identification. First, external ear images are input into the system to detect the visual positive reactions of the auricular points on the external ear using an improved version of the U-Net model. Improvements of the U-Net model include batch standardization, atrous convolution, convolution stage reduction, and multi-expansion-rate integration. Second, the detection results of visually positive reactions are then used to identify the diseases. This study identified nine types of diseases through their corresponding positive reactions, including hepatitis, mastitis, cervicitis, prostatitis, frontal headache, migraine, occipital headache, vertex headache, and headache. The dataset used in this study was collected by the authors and called the CVIU 108 EAR Dataset. The experimental results show that the disease diagnosis accuracy rate of the proposed system is 99.60%.

DOI： 10.1109/LifeTech53646.2022.9754793

Other Link： https://dblp.uni-trier.de/db/conf/lifetech/lifetech2022.html#ChiangHFY22

Automatic Visual Monitoring System for Data Center Device Management

Misheel Enkhbaatar, Tatsuya Yamazaki

WSCE 2022 - 2022 5th World Symposium on Communication Engineering 22 - 25 2022

Publishing type：Research paper (international conference proceedings)

One of the core responsibilities of a data center is to provide stability and availability for online services. But the data center frequently faces infrastructure failures and unpredicted irregularities. In order to minimize hardware related failures, data centers must monitor the housing devices for 24 hours and 365 days. Due to various physical and software related failure causes, monitoring has always been a challenging task. A visual inspection by human operators is a reliable method which is widely used for checking device's status. However, growth of number of devices in data centers causes exponential cost growth for visual checks and it is almost impossible to detect anomaly occurrences in real time. Therefore, we have proposed an automatic visual monitoring system for data centers to substitute human visual monitoring. In this paper, we present the overview of the proposed system and show experimental results to evaluate LED detection performance of the system.

DOI： 10.1109/WSCE56210.2022.9916030

Impact on Sleep Quality of Electronic Device Usage.

Tatsuya Yamazaki

17th IEEE International Conference on Computer Sciences and Information Technologies(CSIT) 2022-November 203 - 205 2022

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

A recent research result reports that the use of electronic devices is one of the factors affecting sleep. Polysomnography, in which EEG (Electroencephalogram), heart rates, and so on are measured, is often used to measure the state of sleep. Polysomnography burdens subjects because electrodes need to be attached to their heads and chests all night long. On the other hand, wearable devices have recently become popular as small and lightweight measurement devices worn on the body due to electronic circuit downsizing. A typical example is a wrist-watch type device, which can measure heart rates and sleep status just by being worn on the wrist. This study aims to clarify the effect of using electronic devices before bedtime on the quality of sleep through experiments. The sleep state of the subjects was measured by using a wearable device in the experiments. The measured results are compared based on the three kinds of tasks to the subject in pre-sleep time: regular books, electronic books (e-books), and electronic games (e-games). The results of statistical analysis for the measured data have shown that e-books and e-games increased the time to fall asleep and decreased sleep efficiency compared to reading regular books.

DOI： 10.1109/CSIT56902.2022.10000506

Other Link： https://dblp.uni-trier.de/db/conf/csit/csit2022.html#Yamazaki22

Dependency of Audio and Video for Two Types of Online Dialogue Conference Services Reviewed International journal

Tatsuya Yamazaki, Shunta Sasaki, Takumi Miyoshi

International Conference on Emerging Technologies for Communications (ICETC2021) ( P4-18 ) 2021.12

Language：English Publishing type：Research paper (international conference proceedings) Publisher：The Institute of Electronics, Information and Communication Engineers

Usually both video and audio media are used for online conferences. Although previous works, in general, support existence of video media because of smooth communication, some research results in no significant differences between video and audio (V-A) conferences and audio-only (A-O) conferences. This work conducts a comparative experiment between V-A and A-O conferences. In the experiment, two tasks are devised: the association dialogue task and the alternative play script reading task. The former is categorized as a contemplation dialogue and needs more discussion and thoughts for the subjects, while the latter is categorized as a light dialogue. From the 14 subject experiment results, video media could contribute to easiness of task achievement. But the contribution was smaller for the play script reading task, because this task uses more superficial communication than the association dialogue task.

DOI： 10.34385/proc.68.P4-18

Quality of Experience (QoE) Studies: Present State and Future Prospect.

Tatsuya Yamazaki

IEICE Transactions on Communications 104-B ( 7 ) 716 - 724 2021.7

Publishing type：Research paper (scientific journal)

With the spread of the broadband Internet and high-performance devices, various services have become available anytime, anywhere. As a result, attention is focused on the service quality and Quality of Experience (QoE) is emphasized as an evaluation index from the user's viewpoint. Since QoE is a subjective evaluation metric and deeply involved with user perception and expectation, quantitative and comparative research was difficult because the QoE study is still in its infancy. At present, after tremendous devoted efforts have contributed to this research area, a shape of the QoE management architecture has become clear. Moreover, not only for research but also for business, video streaming services are expected as a promising Internet service incorporating QoE. This paper reviews the present state of QoE studies with the above background and describes the future prospect of QoE. Firstly, the historical aspects of QoE is reviewed starting with QoS (Quality of Service). Secondly, a QoE management architecture is proposed in this paper, which consists of QoE measurement, QoE assessment, QoS-QoE mapping, QoE modeling, and QoE adaptation. Thirdly, QoE studies related with video streaming services are introduced, and finally individual QoE and physiology-based QoE measurement methodologies are explained as future prospect in the field of QoE studies.

DOI： 10.1587/transcom.2020cqi0003

Machine Learning Approach to Decision Making for Insulin Initiation in Japanese Patients With Type 2 Diabetes (JDDM 58): Model Development and Validation Study. International journal

Kazuya Fujihara, Yasuhiro Matsubayashi, Mayuko Harada Yamada, Masahiko Yamamoto, Toshihiro Iizuka, Kosuke Miyamura, Yoshinori Hasegawa, Hiroshi Maegawa, Satoru Kodama, Tatsuya Yamazaki, Hirohito Sone

JMIR medical informatics 9 ( 1 ) e22148 2021.1

Language：English Publishing type：Research paper (scientific journal)

BACKGROUND: Applications of machine learning for the early detection of diseases for which a clear-cut diagnostic gold standard exists have been evaluated. However, little is known about the usefulness of machine learning approaches in the decision-making process for decisions such as insulin initiation by diabetes specialists for which no absolute standards exist in clinical settings. OBJECTIVE: The objectives of this study were to examine the ability of machine learning models to predict insulin initiation by specialists and whether the machine learning approach could support decision making by general physicians for insulin initiation in patients with type 2 diabetes. METHODS: Data from patients prescribed hypoglycemic agents from December 2009 to March 2015 were extracted from diabetes specialists' registries, resulting in a sample size of 4860 patients who had received initial monotherapy with either insulin (n=293) or noninsulin (n=4567). Neural network output was insulin initiation ranging from 0 to 1 with a cutoff of >0.5 for the dichotomous classification. Accuracy, recall, and area under the receiver operating characteristic curve (AUC) were calculated to compare the ability of machine learning models to make decisions regarding insulin initiation to the decision-making ability of logistic regression and general physicians. By comparing the decision-making ability of machine learning and logistic regression to that of general physicians, 7 cases were chosen based on patient information as the gold standard based on the agreement of 8 of the 9 specialists. RESULTS: The AUCs, accuracy, and recall of logistic regression were higher than those of machine learning (AUCs of 0.89-0.90 for logistic regression versus 0.67-0.74 for machine learning). When the examination was limited to cases receiving insulin, discrimination by machine learning was similar to that of logistic regression analysis (recall of 0.05-0.68 for logistic regression versus 0.11-0.52 for machine learning). Accuracies of logistic regression, a machine learning model (downsampling ratio of 1:8), and general physicians were 0.80, 0.70, and 0.66, respectively, for 43 randomly selected cases. For the 7 gold standard cases, the accuracies of logistic regression and the machine learning model were 1.00 and 0.86, respectively, with a downsampling ratio of 1:8, which were higher than the accuracy of general physicians (ie, 0.43). CONCLUSIONS: Although we found no superior performance of machine learning over logistic regression, machine learning had higher accuracy in prediction of insulin initiation than general physicians, defined by diabetes specialists' choice of the gold standard. Further study is needed before the use of machine learning-based decision support systems for insulin initiation can be incorporated into clinical practice.

DOI： 10.2196/22148

Continuous Experiments for Elderly Person's Exercise Sensing and Evaluation.

Tatsuya Yamazaki

3rd IEEE Global Conference on Life Sciences and Technologies(LifeTech) 341 - 343 2021

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

To motivate elderly people to continue exercises and support busy staffs in care facilities, a sensing and evaluating system is developed for the elderly people exercises. The system consists of a Kinect sensor and a personal computer and detects joint movements of the elderly persons. 5 days continuous experiments have been conducted to evaluate the developed system with the cooperation of a care facility. A male in his 70’s and a female in her 90’s were the experimental subjects. In the experiments, the weather condition data such as temperature and humidity were collected to evaluate the relationship between the weather condition and the joint movement. Analysis of the experimental results reveals tendency of each person’s exercise. Moreover, it is found that atmospheric pressure may affect joint movements.

DOI： 10.1109/LifeTech52111.2021.9391851

Other Link： https://dblp.uni-trier.de/db/conf/lifetech/lifetech2021.html#Yamazaki21

Development of an Automatic Visual Monitoring System for Data Center Devices.

Misheel Enkhbaatar, Tatsuya Yamazaki

IEEE 16th International Conference on Computer Sciences and Information Technologies 2 194 - 197 2021

Publishing type：Research paper (international conference proceedings) Publisher：IEEE

The main responsibility of a data center is sustaining stability and availability for online services. Even so, any data center frequently faces infrastructure failures and unpredicted irregularities. Most of the data centers hosts hundreds or thousands of physical devices. In order to minimize hardware related failures, data centers must monitor the housing devices for 24 hours and 365 days. Due to physical and software related numerous failure causes, monitoring has never been an easy task. A visual inspection by human operators is a reliable method which is widely used for checking device's health. However, growth of number of devices in data centers could cause exponential cost growth for data center management as it requires human work on each LED's inspection, and it is almost impossible to detect anomaly occurrences in real time. Therefore, we propose an efficient and low-cost method for data centers to substitute human visual monitoring. In this paper, we discuss designing and development of automatic visual monitoring system for data centers. Therefore, usability of the automatic monitoring system is experimented in real data center environment.

DOI： 10.1109/CSIT52700.2021.9648588

Other Link： https://dblp.uni-trier.de/db/conf/csit/csit2021-2.html#EnkhbaatarY21

Brain activity sensing to verify effects of facial image reconstruction puzzle—Verification by young students

Shu Iwata, Tatsuya Yamazaki

Sensors and Materials 32 ( 11 ) 3567 - 3580 2020.11

Publishing type：Research paper (scientific journal)

We have developed an application for cognitive function training in order to cope with the increased number of dementia patients and the shortage of care workers in an aging society. This application generates an image in which a person’s facial image is transformed into a puzzle where the aim is to reconstruct the face. The puzzle is presented as a problem of person estimation to activate brain functions. In this study, we conduct two brain activity sensing experiments using electroencephalograms to evaluate the effect of the facial image reconstruction puzzle on cognitive function. The experiments suggest that the image reconstruction puzzle may affect the frontal lobe.

DOI： 10.18494/SAM.2020.2883

A cognitive system of elderly exercise evaluation with sensors and robots

Tatsuya Yamazaki

Cognitive Informatics, Computer Modelling, and Cognitive Science: Volume 2: Application to Neural Engineering, Robotics, and STEM 49 - 63 2020.1

Publishing type：Part of collection (book)

As the aged society is approaching soon, importance of physical practice is increasing for elderly people more to keep their joint flexibility and cognition ability. Although such physical practice is conducted at care facilities in many cases, necessary care is not always available because the elderly population growth is quite rapid. Therefore it is important to introduce ICT (information and communications technology) into the elderly care support at the care facilities to supplement the care shortage. As the first step of introduction of ICT, we propose a cognitive system to watch elderly persons during exercising and to evaluate their exercise. The system consists of three processes. In the first process, body joint data of an elderly person are collected by a depth sensor. Then we extract features to evaluate the elderly person’s exercise from the measured data in the second process. Finally, in the third process, the evaluation is feedbacked to the elderly person by using an intimate robot interface. The cognition system can be extended by using multiple depth sensors to mitigate occlusion. When multiple depth sensors are used, there appear two problems: frame synchronization and calibration. We also approach these two problems.

DOI： 10.1016/B978-0-12-819445-4.00003-5

　 PREV - NEXT 　

1
2
3
4
5
6