In this paper, we propose a technique for automatic scoliosis detection from moire topographic images. Normally the moire stripes show symmetry as a human body is almost symmetric. According to the progress of the deformity of a spine, asymmetry becomes larger. Numerical representation of the degree of asymmetry is therefore useful in evaluating the deformity. First, displacement of local centroids and difference of gray values are evaluated statistically between the left- and the right-hand side regions of the moire images with respect to the extracted middle line. We classify the moire images into two categories i.e., normal and abnormal cases from the features, employing discriminant analysis. An experiment was performed employing 1,200 moire images and 84.72% of the images were classified correctly.

View full abstract

Purpose of medical checkup is early discovery and the treatment of various kinds of diseases. The checkup including discovery of life-style related diseases is carried out annually in companies or communities in Japan. The examination items are almost common, however, under the present conditions, so long as a subject watches values of the result, he or she can merely judge whether the test values are normal or not. It is almost impossible for the subject without knowledge of medicine to judge his or her own health state and to understand what kind of habit must be improved. In this research, the construction of a health evaluation system which visibly indicates a health state was attempted from each person's checkup list by using the feature of self-organizing map (SOM: Self-Organizing Maps) that multi-dimensional data can be mapped to the two-dimensional surface.

View full abstract

This paper deals with a prognosis of patients with gastronic cancer. We employ a heuristic fuzzy clustering and a fuzzy rule-base method on the activity of PyNPase in normal and cancerous tissue for the prognosis of these cancers. By the heuristic clustering based on clinical measurement data concerning differentiation of PyNPase activity between normal and tumor regions, the fuzzy diagnosis rules are designed. The prognosis of patients with cancer is done by these fuzzy diagnosis rules.

View full abstract

The author reported previously that we produced a tongue diagnosis supporting system by applying the syndrome determination process on a discriminating system based on the fuzzy theory. To evaluate the diagnostic ability of our tongue diagnosis supporting system, the results of both the tongue diagnosis and the four methods of diagnosis were compared with those of the diagnosis using this system. The diagnosis using the supporting system was consistent with the tongue diagnosis in 80% and the four methods of diagnosis in 40%. Further detailed studies must be made improving this diagnostic supporting system. From now on, an improvement of this tongue diagnostic supporting system need to add to the diagnostic supporting system of the four methods of diagnosis i.e. inquiry and pulse feeling palpation.

View full abstract

This paper describes Dynamic Knowledge Collection System Using Web Technology. The developed system using the web technology and XML is aimed to dynamically collect a certain significant knowledge in the stile of fairly meaningfully structured, from the miscellaneous dairy memorandum. The several kinds of knowledge collected in this way will support the work to construct higher level knowledge or decision making work, especially in the field of medical activities.

View full abstract

A fuzzy node fuzzy graph is the extension of a fuzzy graph, and it has been applied to the social structure analysis, the instruction structure analysis and so on. The fuzzy node fuzzy graph is usually complicated, so, we would propose a transform method a fuzzy node fuzzy graph to a crisp node fuzzy graph by using T-Norm. But, the crisp node fuzzy graph is also complicated, we would construct its approximate graph. In this paper, we would not only explain about the transformation method and approximation method, but also its practical application to Sociometry analysis.

View full abstract

Although some computer-assisted diagnostic systems have been reported recently, they have not been so much spread in clinical field yet. The principles of fuzzy or neural network are so complicated to the general clinicians and they do not have the skills or techniques, which are necessary to develop the computer-assisted diagnostic system. The authors herein propose the new method, which automatically optimize the system only by setting the simple inferences. Because the method can be utilized with an ordinary application on a standard personal computer, even the users who cannot make programs can develop the system. The method applies the adaptive neuro-fuzzy inference system (ANFIS), which is combined with fuzzy and neural network. In this method, membership functions, which are set temporally at first, can be optimized automatically by back propagation procedure of neural network. Consequently, the performance of the diagnostic system can be improved. The authors describes the outline of the method with the preliminary applied case to the ultrasonographic dignosis.

View full abstract