An effective algorithm for the Hopfield neural network model is proposed through its application to the traveling salesman problem. By the estimation of local minimum inside a hypercube of solution space, a threshold value for deciding integer valued solution can be properly chosen. An optimum solution of the TSP is obtained with very high probability.

View full abstract

Human beings are more subjective than objective. In this paper, paying attention to our internal emotion, especially "subjectivity," we show an application of Frame Oriented Theory to decision processing of psychiatric diagnosis. Most medical diagnoses depend on the subjectivity of its specialist. Occasionally they are highly influenced by their subjectivity. Especially, a subjective diagnostic statement like "it is just like..." by the specialist is necessary in psychiatry. By means of a mathematical model we define the frame for a psychiatric diagnosis. We constructed a psychiatric diagnostic frame, and consider how a patient may be diagnosed by several psychiatrists who have individual subjectivities.

View full abstract

We propose a fuzzy inference ruler using d'Ocaone's nomography. We will be able to get the results of fuzzy inference by the ruler without computer. As the inside of the black box of fuzzy inference will be visible on the ruler, it is expected that the fuzzy theory will be generally accepted and utilized by most persons. In this paper we will first refer to the nomogram, then describe how to apply the ruler to fuzzy inference, and we will discuss about the fuzzy inference ruler made as a trial for deciding exposure parameters of radiography.

View full abstract

In colloid liver scintigraphy, pattern of diffuse parenchymal liver disease such as chronic hepatitis or cirrhosis is evaluated by size and distortion of the liver, distribution of tracer in the liver, size and activities of tracer in the spleen, visualization of the bone marrow and so on. It is not difficult to read a scintigram which shows typical pattern of normal, chronic hepatitis and cirrhosis. But in some cases it is difficult to read normal or chronic hepatitis and chronic hepatitis or cirrhosis in visual diagnosis, so we tried to use fuzzy reasoning to make differential diagnosis in chronic hepatitis (CH), bridging fibrosis (BF) and cirrhosis (LC). First, five features in colloid liver scintigrams were evaluated visually. These features were left lobe/right lobe, splenomegaly, visualization of the bone marrow, liver deformity, and distribution of tracer in the liver. Having fuzziness in these data, tendency of these features, 'small', 'medium', 'large' for instance were thought fuzzy sets and displayed in membership functions. Fuzzy reasoning was carried out using these data and fuzzy rules. Using fuzzy reasoning differential diagnosis in LC could be done in 100% ,but CH and BF could not be differentiated. So using neural network CH,BF and LC could be differentiated, therefore combining fuzzy reasoning and neural network better differential diagnosis will be expected.

View full abstract

This paper introduces basic concept of an emotional dialogue system and shows how to construct the system by computer simulation. We discuss a dialogue between 2 simulated persons who know only emotional words. At first, we show Image Code Table (ICT) of mixed emotions which are regarded as emotions in usual life, introduced by R. Plutchik. Secondly, we also show how to design an emotional dialogue system based on the theory for Subjective Observation Model (SOM), which have high applicability to several kinds of fields. We explain the outline of the theory. Finally, we show several attractive dialogues by computer simulation.

View full abstract

In this study, I propose Neural Network, Fuzzy System, and Multiple Regression models that use visual inspection data directly to provide the estimations. I had designed a dental expert system for the evaluation of dental developmental age, based on computer entry of data on the eruption status. The resuls showed the almost same averaged squared error. But estimation results of multiple regression and neural network were better than beginner of pediatric dentist. And, the lower performance of the fuzzy system models may be due to an insufficient number of principal components used in the input space or parameters of the learning algorithm.

View full abstract

In this paper we propose the fuzzy artificial intelligent system with learning function, which is called "FAIS" short for. This system has some functions which acquire new fuzzy knowledge from experiences under an unknown situation, and can solve some problems by deducing new state through fuzzy reasoning. Also, the purpose of constructing this system "FAIS" is to make a system that can solve many kinds of problems in various fields. As an example, our system "FAIS" is applied to a medical diagnosis and treatment advice system, and the usefulness of this system is discussed from simulation results.

View full abstract

We have established a fuzzy set cloning to identify histo-regulatory protein. For the first step, we have screened cDNA library using antibody against basement membrane. For the defuzzyfication we checked obtained cDNA clones (32 positives) on Northern blots using purified mRNA obtained from fibrogenic liver of rats. One of the clone activelytranscribed at 3 hrs after CCl4 injection was sequenced, and identifiedas profilin cDNA.

View full abstract

It would be more practical to represent the human relations using a fuzzy graph rather than a crisp form such as a hard cluster. According to the data obtained from several questionairs, we could measure the connectivity degree among members and construct its fuzzy graph. Applying the fuzzy clusterings and orderings, we have developed an analysis method of a sociogram which was proposed by Moreno. In this paper, we would not only discuss our analysis method, but also show its practical effectiveness through the case study.

View full abstract

This paper describes a model of a consciousness field in animal behavior. This model, originally proposed by Tao from a phenomenological viewpoint, is based on an idea that a consciousness appears when an intentional behavior is stopped for any reasons. Applicability of this model is theoretically investigated with use of an imaginary animal. Feasibility of the model is concluded for downloading on an autonomous robot simulating animal's behavior.

View full abstract

Recently, number of applications of fuzzy theory to medical analysis and modeling is increasing especially in western medicine, but the applications in eastern medicine could almost not be found. Then this paper discusses how the approach based on fuzzy theory and the theory of subjective observation model is also applicable to eastern medical modeling.

View full abstract

The influences of physiological and psychological factors should be taken into account for accurate evaluation of ride-quality. Psychological effects of 165 subjects on board a ship-motion simulator were measured by means of questionnaire survey. A mathematical model using fuzzy measure theory was proposed to analyze the psychological change of subjects during exposure to low-frequency whole-body oscillation.

View full abstract

Esthetic improvement of the face is one of the goals of orthognathic surgery. When attempting to make an objective evaluation of the profile as a result of the surgery, soft tissue measurements on a lateral cephalometric radiogram are currently used, which are compared with the norm. However, no single measurement provides an overall evaluation of the profile esthetics. Contrarily a subjective impression by eye gives a comprehensive evaluation, although there is a problem which can not be quantified due to its intrinsic fuzziness. Then we applied the fuzzy reasoning to modelling a subjective evaluation of profile and have developed a computer system (prototype).

View full abstract

The line drawing is generated from input image by using neural network as feature detection. The primitive pattern is drawn on the output image according to the detected features. The local line segment is detected as the feature. The various ouput image is generated by changing the some parameters.

View full abstract

This paper explains a try to realize an associative system from a certain figure to the related colors or tastes,by the way of linear combination and affine mapping.

View full abstract

Resently, with the development of the technique of computers lot kind of the research in the field of facial psychology has been showed. In this paper, we will introducethe theory of the subjective observation model at first, then do some experiments based on the theory. The purpose of this research is to make the subjective and objective information procession about the feeling of human being that expresed from face, and decide the pattern of it.

View full abstract

This approach is one that separates the input space of variables of fuzzy rules (model) into some subspaces, and unifies sub spaces step by step. In this paper, we propose the intelligent clustering for fuzzy modeling. Then we show results of a numerical example and an application to clinic.

View full abstract

This paper proposes a new method to solve a diagnostic problem using fuzzy abductive inference. The fuzzy logic is applied to Peng and Reggia's abductive inference for handling degrees of manifestations. Application of the new method to a simple diagnostic problem is shown and the effectiveness of the proposed method is demonstrated by comparing the inferred results with those by the conventional fuzzy abduction.

View full abstract

AIDS (Acquired Immunodeficiency Syndrome) is now increasing all over the world, and it gives a new kind of social crisis to every countries. We simmulate the epidemiology of AIDS and carier of HIV (Human Immunodeficiency Virus) by using fuzzy set. Two major membership factors such as effect of education and effectiveness of propagation were input. The computer program using fuzzy theory predicted long period of AIDS disaster as if the "National tragedy" would be the truly occurring event.

View full abstract

Human growth in the relative lower limb length (= a ratio of lower limb length to stature) is analysed using logistic curve. The human growth is the increase of the cell. The cell divides into two new cells, but the deaths of the cells occur afterwards. The growth of the relative lower limb length curve runs almost diagonally until 14 years old, and almost horizontally along the asymptote after that age. The curve of the mean relative lower limb length in the present study seems to show the maximum 46.1 at the highest. The constant of the logistic curve is estimated at 1.78 < a < 1.86.

View full abstract

Objective of this paper is to find capabilities to realize a face animation system, combining the emotion processing system developed in our laboratory in witch the investigated useful results concerning the mixed emotions by the well- known psycholosist, R. Plutchik are taken into accout. The face expression is done by the combination of the movements of eyes.eyebrows and a mouth.And each of the movements is drawn by Bezier function whose control points can be extracted from the resultant image codes after the emotional processing.

View full abstract

A fuzzy recognition system is used here as a medical expert system for diagnosing hepatic diseases. This system automatically decomposes the total feature space into some overlapping subspaces and accordingly finds a few ifthen rules. The performance of the system is found to be better than all other methods applied on the same data set for identifying hepatic diseases.

View full abstract

Frcatal theory is an effective method to describe quantitatively the complicated shape in nature. We applied the fractal theory to medical images. We have utilized scale conversion method to obtain fractal dimensions. A water phantom was scanned with CT under various conditions. The CT images were converted to digital data with pixels of 256×256 and densities of 6 bits. The results showed fractality even after many steps of conversion. The fractal dimensions decreased with decrease of photon noise. Fractal dimensions were thought to be applicable for the quantitative evaluation of CT images.

View full abstract

1/f-Fluctuations is widely observed in various environment, such as in a breeze, in a sound (music), in a smell, in a pulse of creature and so on. For long time, the fluctuation had been though to be a mere noise. But recently, it had recognized that the fluctuation brings us comfortable. In this paper, we describe the relation between 1/f-fluctuations and comfort, and discuss how to implement the fluctuation generator on the specific hardwere. Then, we suggest an approximate l/f-fluctuations controller by chaotic oscillator.

View full abstract

New fuzzy reasoning Method called "fuzzy singleton-type reasoning method" is introduced which treats fuzzy rules whose consequent part is a fuzzy singleton (that is, a real number with a positive weight). It is shown that product-sum-gravity method which deals with ordinary fuzzy rules whose consequent part consists of a fuzzy set is reduced to the fuzzy singleton-type reasoning method by regarding the area of a fuzzy set of the consequent part as the weight of the fuzzy singleton-type reasoning method. Therefore, it is found that fuzzy singleton-type reasoning method is equivalent to product-sum-gravity method.

View full abstract

Diabetes mellitus is one of the typical adult disease, whose patients number will arrive at six millions in Japan. Foods control and exercises take an important role for the treatment of diabetes mellitus. However, this self-control for diabetes is depended upon the efforts of patients and their families. Therefore, the treatment of foods control and exercises must be controlled by the communication between physicians and patietnts. In this paper we proposed a supporting system for the self-control of disbetes mellitus with the communication system between physicians and patients using a fuzzy inference based on the daily data such as the data of blood sugar, foods, exercises, the degree of stress, and so on. We applied this system to the patients' data of diabetes and discussed this system from the clinical and computational points of view. From the results of its application, it was suggested that this system was useful for the clinical communication system of home care through the multimedia information highways.

View full abstract

The projection pursuit (PP) for regression was originally proposed by Friedman and Stuetzle. The important feature of PP is that it is one of the multivariate methods able to bypass the "curse of dimensionality". The aim of PP is to find an interesting or characteristic structure by working in low-dimensional linear projections. In this paper, we propose a method based on the membership function and the eigenvector of the covariance matrix to avoid the local minimum of the projection indices.

View full abstract