A new method is addressed on the 3-D modeling of human postures/motions based on multiple cameras. The proposed method doesn't perform camera calibration before image taking. It is therefore applicable to 3-D modeling of unplanned or unexpected events. In the next topic, a method of human motion recognition is shown which employs appearance-based approach. In the method, every possible appearance of a human posture/motion is registered in a memory, and the recognition is realized by retrieving it. Since the registration is done on a structured memory, effective retrieval is achieved in the method.

View full abstract

Developing a Clinical Reasoning System for medical decision-making in the clinical field is important in order to improve accuracy of the judgment of clinician, and to reduce risk of the diagnosis error. This paper proposes a Clinical Reasoning System based on clinician's diagnosis models. This system applies inverse posibilistic causal reasoning using uncertain disease-symptom knowledge to conclude the disease from the measured symptoms.

View full abstract

We compared the brain activation during both playing a 3D action video game and viewing the game. The two tasks activated visual association cortex, superior parietal lobule and cerebellum. The game playing activated dorsal prefrontal cortex, anterior cingulate cortex and thalamus. The result indicated that playing the game caused attention and emotion.

View full abstract

Old drivers on vehicles of cars, trains and airplanes, etc. are increasing in our aged society advancing today. Therefore it's expected that the probability occurred of a serious accident by incapacitation of an old driver will be high. The analytical result of ECG is indispensable as the indicator to stop accidents of this kind from happening. The purpose of this paper is to estimate the model to solve general nonlinear predictability problems taking the case of an ECG. The models to evaluate comparison are a non-recursion type 1st-order Volterra neuron network (N1VNN) and a non-recursion type Volterra neuron network (NVNN) of discrete time. The result is which the NVNN is better than the N1 VNN.

View full abstract

In recent years, various imaging techniques have been introduced into medical fields for detecting of abnormalities such as cancer, shadows. Especially, HRCT.is one of the useful diagnosis systems because it provides high resolution images. In the field of abdominal CT images, it is very important to understand blood vessel structures of a patient before operation. On the other hand, temporal subtraction method attempt to remove normal structures in the CT images, so that abnormalities can be observed more clearly to the medical doctor. With the subtraction techniques, image warping is very important. In order to achieve a good performance, elastic matching technique is applied to temporal subtraction method. In this paper, we apply temporal subtraction method with elastic matching technique to abdominal CT images in order to detect the blood vessel region for analyzing the internal organ. In this method the temporal subtraction is applied between successive phases. The proposed technique is applied to three abdominal CT images and satisfactory results are achieved.

View full abstract

Automatic detection of abnormal shadow area on a multi detector CT image is important task under developing a computer aided diagnosis system. Ground glass opacity is one of the important features in lung cancer diagnosis of computer aided diagnosis. It may be seen as diffuse or more often as patchy in distribution taking sometimes a geographic or mosaic distribution. A large number of diseases can be associated with GGO on CT image. We propose a technique for automatic detection of ground glass opacity from the segmented lung regions by computer based on a set of the thoracic CT images. In this paper, we segment the lung region for extraction of the region of interest employing binarization and labeling process from the inputted each slices images. The region having the largest area is regarded as the tentative lung regions. Furthermore, the GGO is classified by using density features on the successive slice from the extracted lung region with respect to the thoracic CT images. Experiment is performed employing 32 thoracic CT image sets and 71.7% of recognition rates were achieved. Some experiment results are shown along with discussion.

View full abstract

In this paper, we propose a new automatic classification method for spinal deformity from moire topographic images which obtained asymmetric degree by using a new statistical feature. In our method, we detect three features such as mean values, standard deviation from an original moire image and cross correlation value from a converted low resolution image. Extracted three features are applied to employing an artificial neural network for classification. The proposed technique applied 1200 real moire topographic images and obtained some experimental results are shown along with a discussion.

View full abstract

The purpose of this study is to assist the activities of daily living (ADL) of disabled persons. In our system, surface electromyogram (EMG) signals, contaminated by electrical stimulation, are measured to control the movement in ADL. Double differential amplifier (DD) and blanking are used as noise reduction techniques. We developed the system by using a MATLAB based DSP, which is suitable to process the EMG and separate artifact signals in real time. A simple movement experiment of grasping an object with the hand and lifting the forearm was done successfully.

View full abstract

It is well known that musculoskeletal atrophy arises among astronauts and bedridden patients. It seems that the disuse atrophy of the muscles and bones is acute in low gravity environments. The atrophy is one of main issues for space medicine and clinical medicine. We proposed a "hybrid exercise method" (HYB exercise) in which antagonist is stimulated electrically to generate resistance forces against volitional contraction of agonist. We confirmed its advantages, but found it be tedious in a long-term exercise. The motivation for the exercise may be decreased by repeated monotonous movements every day in a long-term exercise. The purpose of this study is to develop a useful training system for maintaining or increasing motivation to exercise. A proposed VR training system will give relief from the mental stress and the atrophy in weightless environment.

View full abstract

Tracheal intubation is the standard medical act to conduct airway management for artificial respiration. A laryngoscope and a tracheal tube are always used in that medical act. The way to prevent a complication while using a laryngoscope is only compassionate treatment from effectors. There are no objective guidelines of that medical act. When tracheal tube is inserted into the trachea to send air, it happens to be inserted into the esophagus by mistake. The incorrect intubation is very dangerous. In this paper, force sensor and two-axis gyroscope sensor are attached to the laryngoscope for making objective evaluations, and alarm buzzers beep to warn wrong laryngoscope actions. A field test was performed by using a training model, and the result was analyzed by principle component analysis to evaluate the degree of achievement and a personal characteristic. In addition, acoustic signals from a tracheal tube were examined as to whether insertion was correct or incorrect. Adaptive filter was used to reduce noises in the signals for enhancing estimated precision.

View full abstract

Spinal deformity is one of a serious disease mainly suffered by teenagers during their growth stage. To detect the spinal deformity in early stage, orthopedists have performed a painless examination which called a forward bending test or moire topographic image test in mass screening. It is, however, inspection base on the forward bending method takes much time and the moire images and visual examination also require a large amount of moire images because from elementary as well as junior high schools are collected. This causes exhaustion of orthopedists and therefore leads to misjudgment. Therefore realization of automated scoliosis inspection based on the moire images has long been desired among orthopedists. In this paper, we propose a method for automatic detection of spinal deformity from moire topographic images by using a new asymmetric feature which is obtained difference of left- and right-hand side shoulder and waistlines. The proposed technique applied 1200 real moire topographic images. Obtained some experimental results are shown along with a discussion.

View full abstract

A clear digital image of blood vessels by using a CT angiography or MR angiography is easily obtained in medical field. Recently, the disease by arteriosclerosis increases rapidly. Especially, the one that originates in hands and feet's peripheral arterial is called arteriosclerosis obliterans (ASO). To diagnose the ASO using a magnetic resonance imaging (MRI) applied as a useful visual screening method on the medical field. To avoid adverse reactions from the contrast medium non-enhanced fresh blood imaging (FBI) method is developed. In this paper, we develop a 3-D visualization method for analyzing the internal structure such as blood vessels in the lower limbs area. To develop a visualization method, we propose a technique for extracting the blood vessels region by use of region growing method which obtained the MIP image. This technique was applied to four non-enhanced FBI MR imaging, and satisfactory 3D blood vessel region was extracted. Some experimental results are shown with discussion.

View full abstract

Detection of abnormal areas such as lung nodule, ground glass opacity (GGO) on multi detector computed tomography (MDCT) images is a difficult task for radiologists. It is because subtle lesions such as small lung nodules tend to be low in contrast, and a large number of CT slice images required for visual screening times. In order to detect the abnormalities by use of computer aided diagnosis (CAD) system, some technical method have been proposed in medical field. Despite of these efforts, their approach did not succeed because of difficulty of image processing in detecting the GGO areas exactly. Thus they did not reach to the stage of automatic detection employing unknown thoracic MDCT data sets. In this paper, we develop a CAD system for automatic detection of GGO areas from thoracic MDCT images by use of four statistical features. The proposed technique applied 32 thoracic MDCT image sets in the performed experiment, and 77% of recognition rates were achieved. Obtained some experimental results are shown along with a discussion.

View full abstract

In this paper, sales of every day of a certain apparel store were investigated for the consumer's demand forecasting model construction, and the meteorological factor such as the temperature was investigated at the same time. And, consumer' s demand forecasting model is constructed by using meteorological factor in statistically clarifying the relation between these. Concretely, the fuzzy reasoning model to whom demand is forecast by trying the analysis of sales of every day and the relation between the number of visitors and the meteorological factor by applying a decentralized analysis, selecting the meteorological factor from this analytical result, and using the obtained factor is constructed.

View full abstract

This paper proposes a 3D modeling method employing factorization and back projection. The method needs neither 3D camera calibration tool nor markers for shape recovery. From factorization, affine camera model is obtained and back projection employs the affine camera model to recover the original 3-D shape. Experimental results show satisfactory performance of the proposed method.

View full abstract

In this paper, we propose a technique for extracting a human region from images and for recognizing human postures. We employ a background subtraction technique using texture information and gray scale for extracting a human region and a technique for recognizing a 3-D object from a 2-D image. Many extraction techniques must be performed in the outdoors and need thresholds tuned by manual operations and many recognition techniques need extraction of specific characteristics or modeling. Since the proposed technique does not necessitate manual operations, it can perform the recognition from extraction automatically. The proposed technique obtained satisfactory results by the experiment performed in the outdoor environments.

View full abstract

This paper describes a fast 3-D reconstruction technique using two cameras. The existent 3-D reconstruction techniques using two or more cameras take a lot of processing time. The proposed fast 3-D reconstruction technique simplifies a past 3-D reconstruction technique using epipolar geometry and a camera rotation matrix, and employs the OpenCV for fast matrix computation. The proposed technique shows satisfactory performance in the performed experiment.

View full abstract

In this paper, we develop a robot which can detect and track a human face for the purpose of developing an autonomous robot that communicates with people. In order to detect a human face from a camera image, we pay attention to flesh color of a person and develop a technique for detecting and pursuing a flesh-colored area as a face. The developed robot simulates a human head. We install two USE cameras on the part which are eyes of the robot head. The robot head can perform pan-tilt movement by two servomotors. In the performed experiment, the face of a person walking around in front was detected and tracked successfully by the developed robot head.

View full abstract

Today E-learning is extending because people came to use a personal computer and the Internet, Then, a lot of contents is demanded, but past contents are insufficient to train the researcher. It is important to draw out student's inquiring mind. Then This Paper describes the trial about E-learning method with the study process to draw out student's independent ambition and the interest in technique.

View full abstract

This paper deals with analysis of cluster formations on planer cells consisting of multi-agents using local interactions and state transitions based on the Genetic Programming and its applications. It is shown if the states of agent are restricted to be binary, then the equilibrium is attained only in an asymptotic sense. However, in cases where the number of states is larger than one, the attainment of equilibrium is not ensured. Then, we examine the result by using the simulation studied where agents learn form past experiences based on the GP.

View full abstract

This paper relates the effects of repeated listening to "One" wind band music for heart rate variability (HRV) in healthy woman who received music education. One woman (aged 31) who received music education was invited as the experiment subject, and one piece of wind band music ("Appalachian overture" (composed by J. Barnes)) is employed as the test piece. The subject is asked to listen to test piece for 2weeks (one time per a day), and her HRV is recorded. The experimental data is analyzed by FFT, components of low frequency (LF ; 0.04-0.15Hz) and high frequency (HF ; 0.15-0.40Hz) are obtained from the analysis of power spectrum in HRV. As the number of listening to the test piece is increased, the changes of LF/HF ratio are observed.

View full abstract

We drew an analogy between a quantum and a subjective datum, and applied the quantum theory to a subjective data analysis. As a result, the application to fuzzy equivalence relation presented the availability of fuzzy clustering.

View full abstract

Objective : Today in the United States, about one in eight women over their lifetime has affected by breast cancer. In recent years the incidence rate keep increasing and data shows that the five years survival rate is 88% and the 10 years survival rate is 80%. Recently in Japan, the incidence rate of breast cancer has been increasing. After 1995, breast cancer ranks first incidence in female cancers and receives a great deal of public attention. Up to today, various prediction models with using SEER (Surveillance Epidemiology and End Results) datasets have been constructed in past studies. However the appropriate methods for predicting the breast cancer have not established. In this paper, we present optimal models to predict the survival rate of breast cancer patients in five years. Material and Methods' This study used the 37,356 follow-up patients by 2002 that were diagnosed as breast cancer and registered in the SEER program from 1992 to 1997. We used seven common data mining algorithms (Logistic Regression model, Artificial Neural Network (Multilayer Perceptron), Naive Bayes, Bayes Net, Decision Trees with naive Bayes, Decision Trees (ID3) and Decision Trees (J48), besides a most generally used statistical method (Logistic Regression model) to develop the prediction models. We also used 10-fold cross-validation methods to assess the unbiased estimate of the five prediction models for comparison of performance of each method. Results : The accuracy was 85.8±0.2%,84.5±1.4%,83.9±0.2%,83.9±0.2%,84.2±0.2%,82.3±0.2%,85.6±0.2% for the Logistic Regression model, Artificial Neural Network(Multilayer Perceptron), Naive Bayes, Bayes Net, Decision Trees with naive Bayes, Decision Trees(ID3) and Decision Trees(J48), respectively. Conclusion : In this study, Logistic Regression model showed the highest accuracy. The Decision Trees (J48) had the highest sensitivity and the Artificial Neural Network had the highest specificity. The Decision Trees models tend to show high sensitivity. And the Bayesian models were apt to show the accuracy goes up. We found that the optimal algorithm might be different by the predicted objects and datasets.

View full abstract

In this paper we propose the exponential fuzzy numbers in Analytical Hierarchy Process (AHP), employing the fuzzy concept. The objective was determined a method of deriving the weights of criteria and their alternatives. Next, the influence of fuzzy is examined by using the membership function. Specifically we use the exponential fuzzy numbers for the pairwise comparison scale. Same experiments with AHP were conducted using the pairwise comparison scale and the new pairwise comparison scale of the method was examined.

View full abstract

Rough set theory was proposed by Z.Pawlak in 1982. This theory can mine knowledge as a decision rule from a database, a web base, a set and so on. The decision rule is used for data analysis as well. And we can reason an unknown object using the decision rule. The objective of this paper is to apply the rough set theory to analysis of time series data. It is possible to acquire knowledge from time series data using regression line and apply a method to predictions..

View full abstract

This Paper describes an Acquisition Method for the Knowledge Based on Action Script to achieve the purpose on some subject. In this research, we tried to develop the system that can issue the keywords automatically according to the script to achieve the purpose. The developed system, especially this time, is aimed to gather the knowledge chips (keywords) automatically using the Acquisition Method based on Action Script from the homepage data made by "The Ministry of Health and Welfare of Japan".

View full abstract

In general, activation of muscles is recorded as electromyography (EMG) with surface electrodes. Because a muscle in our body is composed of motor units (MUs), it is obvious that EMG contains several motor unit action potentials (MUAPs). Also, uncertainty of their mixing ratio makes this task be hard to handle. Then, decomposing EMG into MUAPs is one of the attractive topics today especially in the field of motion analysis. In order to challenge this severe problem, independent component analysis (ICA) is adopted in this study. And, as the first step, some primitive experiments with biomimetic EMG signals are executed to estimate its applicability. As a result, it is found that the proposed method shows good performance even though some problems are revealed.

View full abstract

The rapid dissemination of the World Wide Web raises the issue of the Web accessibility, and one of the important things is the combination of a foreground color and a background color. In our previous study, the visibility of web-safe colors on the white background was examined, and the blue used for the standard link color was found high visibility in wide range of ages. Since the usage of the blue and an underline are recommended as a link, in this study, we examined high-visibility background colors to the standard link color, i.e. blue. One hundred and twenty three background colors to the blue were examined using pair comparison method, and the evaluation of standard link color visibility by CIE-Luv uniform color space. As the result, effective background colors to the standard link color were determined on the CIE-^*Luv, that is, L^* more than 67.6, u' less than 0.258, or L^* more than 67.6, v' larger than 0.421 provided high visibility in wide range of ages.

View full abstract

In this article, we describe about paired comparison test for white background and other character colors selected from the Web Safe Colors, and an analysis of fundamental visibility on Web color sets by SOM (Self-Organizing Map). As visibility, we focused only on the combination of background and character colors, among the design-elements of Web contents. Then we consider a fundamental feature of visibility for the specific age groups using maps generated by SOM. The maps are available for improving visibility and individual personal usability. We are evaluating our web browsing system with an automatic color correction (palette changing) for 〓〓〓〓 use 〓 visibi〓〓, wKicK is b〓ed 〓n 〓〓〓〓〓〓 'Web C〓〓〓 SeW 〓〓〓〓 Method by SOM". Since DuseU profiling data is required beforehand for every user, we solved this problem with SOM.

View full abstract

This Paper describes the trial system that is named partner & server computing system. Partner & Server computing system is the system that records know-how knowledge . know-how knowledge is the personal chip of knowledge ,it doesn't have general administrative information so it can't recorded into usual information retrieval system. It is the problem how to record a personal know-how knowledge,and how to convert Individual know-how information into general knowledge that everyone can use.

View full abstract

Workload depends on each persons working posture and a compelled posture causes their fatigue. In order to improve workload, working postures have been analyzed. But, it is difficult to collect and classify posture data correctly because existing work posture analysis depends on observer's eyes, not expert. In this study, we propose a work posture classification method focusing on the joint angle of workers. Our method classifies posture using fuzzy inference method based on some acquired joint angle data of workers from motion data. Indeterminable postures which overlap between segments of basic posture can be classified by using fuzzy inference method. From experiments, we found that the classification results of a posture had high accuracy.

View full abstract

Recently, "Washer Dryer" type of washing machines with horizontal or slant drum becoming popular in Japan. We have measured and analyzed the posture while using the machines with the 3 dimensional motion capture measurement device. Subjective Kansei and usability questionnaire were also used. After the measurement, measured working postures are analyzed with the human kinematics model (3D SSPP) that can estimate theoretical value of the muscle tension and loads on the lumber vertebrae, knees and ankles. Three types of washers (European type ; box shape and horizontal drum, Conventional Japanese type ; a vertical drum, New type ; slant drum with higher profile) were used for the experiment. Subjects were 12 females and those age were between 22 to 43. Theoretical muscular forces (%MVC ; a percentage of the maximum muscular force of 50 percentile female) on elbow, hip, knee and ankle were estimated. Sum of the %MVC of the new type was 116, European type was 133 (knee tension is high) and conventional Japanese type was 284 (ankle tension is 110, which exceeds the limit). The new type requires only 40% of the muscular force than conventional Japanese type. From the subjective evaluation, the new type was better than the European type washer and conventional type on two of the question, subjective fatigue and general evaluation, with statistical significance with one-way ANOVA.

View full abstract

This paper is concerned with the regression trees used for the analyses of Kansei evaluation data. The methods to learn a regression tree is adopted to Kansei evaluation data which is usually represented as ordinal scale. Two types of method which are based on CART and GAare illustrated in this paper. Actual Kansei data from the experiment for basic designs is used for the numerical tests of the methods.

View full abstract

Project management is one of central issues in management of technology and engineering. Kathy Schwalbe summarized recent statistics that average time overrun is 163% in 2001 and 227% in 1995, and average cost overrun is 145% in 2001 and 189% in 1995. The project management is not so much successful. A software development company is expected to provide high quality and functional software to the world. It is difficult to control software quality because software is invisible and cannot be felt with our finger. In addition, it is hard to make its productivity efficient because software is created as a result of brainwork. Generally speaking, a project team is organized to create software. Human relationships and work environment affect software qualities. Therefore, it is very important to study the project management of software development for the cost of management and the quality control. The objective of this paper is to illustrate the influential features of software development projects by analyzing questionnaires collected from several software development companies. In the analysis, a multivariate model is employed to quantitatively evaluate the influential features of software development projects.

View full abstract

We could generally analyze the inexact information efficiently and investigate the fuzzy relation by applying the fuzzy graph theory. We would extend the fuzzy graph theory, and propose a fuzzy node fuzzy graph. And we transform it to a fuzzy graph by using T-norm family. In this paper, we would discuss about three subjects, (1) fuzzy node fuzzy graph, (2) new T-norm "quasi logical product", (3) decision analysis of the optimal fuzzy graph G in the fuzzy graph sequence {G}. By using the fuzzy node fuzzy graph theory and this new T-norm, we could clarify the relational structure of fuzzy information, and by using the decision of an optimal level on a partition tree, we could analyze the clustering relation among nodes. Moreover, we would illustrate its practical effectiveness with the case study concerning sociometry analysis.

View full abstract

We often use fuzzy graph to analyze inexact information such as sociometric structure, instruction/ cognition sequence, opinion poll and so on. Concerning the hierarchical cluster analysis of a fuzzy graph, it is interesting how to decide the optimal cutting level as to a partition tree. In this paper, we would not only propose the decision method, but also illustrate the typical cluster analysis methods in multivariate analysis and its evaluation method. Then, we proceed to explain the practical effectiveness through the sociometry analysis.

View full abstract

In this paper, we investigate the cause by which study of a neural network stagnates, and propose the technique of raising the success rate of study training. We report using a simulation results that this technique is effective.

View full abstract

In this paper, exchange probability of genes in inversion was described. We categorized the inversion into four types of modes. In mode 1, the gene at the start point in the exchange region has the highest exchange probability than that of other genes. In mode 2, the exchange probability of the genes of near center is higher than that of near the ends. The exchange probability of genes in mode 3 shows tendency in the contrariety with that in modes 2. The exchange probability of the genes in mode 4 has same value. Also we show that the exchange probability of genes near center in mode 1 to mode 4 has same value.

View full abstract

The objective of the paper is to analyze the bio-pharmaceutical industry in Japan, clarify the present state of the bio-pharmaceutical industry and finally quantitatively estimate the relationship of the scale and investment behavior with the management achievement of bio-related companies based on the theory proposed by industrial economist J. S. Bain. That is, the paper is contributed to illustrate a surplus profit and a rate of return of a corporate relating with its capital, an employee size, a general investment and development investment of the corporate. This paper explains how management to enhance the performance of a corporate.

View full abstract