Arata Suzuki

Feature Selection Method for Estimating Systolic Blood Pressure Using the Taguchi Method

This paper proposes a method for selecting photoplethysmogram features for estimating systolic blood pressure (SBP), which uses the signal-to-noise ratio (SNR) obtained from the Taguchi method. Cuffless estimation of blood pressure using just one photoplethysmograph was carried out based on the photoplethysmogram features. However, these features are highly vulnerable to noise generated by sources such as individual variability, resulting from factors such as age and gender. Therefore, features need to be robust against such noise for accurate estimation of blood pressure. In this study, we used an orthogonal array and the SNR from the Taguchi method for selecting features of multiple regression analysis that are robust against noise. We estimated SBP from two datasets by applying the proposed feature selection method to multiple regression analysis. The obtained results show that the correlation coefficients between the actual measurement values and the estimated values were 0.88 and 0.87 for datasets 1 and 2, which are higher than those of similar methods used for comparison purposes. This confirms the effectiveness of the proposed method in estimating SBP.

POLYNOMIAL MATRIX APPROACH TO INDEPENDENT COMPONENT ANALYSIS: (PART II) APPLICATION

Abstract This paper proposes two types of application of the method given in the paper (PART I). The first is suppression of disturbance with unknown dynamics in control systems. Assuming that a disturbance enters into a system independently of the control input signals, the method makes it possible to identify a denominator polynomial matrix of the disturbance dynamics, and then to design an H ∞ controller suppressing the effect of the disturbance. Numerical simulation is carried out to verify the effectiveness of this approach. The second is fault detection of mechanical systems subject to vibration from the environment. The vibration source is assumed to be unmeasurable, and yet the method achieves detection of a parameter change of the system due to fault. To test the effectiveness of the second approach, experiment with a flexible structure has been carried out.

Robust estimation of heat capacity in cooking devices

This paper proposes a method for the estimation of heat capacities in cooking devices, particularly rice cookers. In this method, the response to ON-OFF control of the cooking device is used as characteristic data, and the heat capacity is estimated in terms of a minimal cooking unit in the characteristic data space. The ON-OFF control is based on a single threshold irrespective of the heat capacity to be estimated. Hence the response changes with the heat capacity, but it is difficult to estimate only from the response, because the estimate is sensitive to the initial state. The proposed method collects multiple characteristic data, then selects only the ones effective for the estimation, and obtain their weighted average, so that it achieves robust estimation. The effectiveness of the proposed method is verified through experiment with a rice cooker.

Inverse-model-based cuffless blood pressure estimation using a single photoplethysmography sensor

This paper proposes an inverse-model-based cuffless method for estimating blood pressure using a single photoplethysmography sensor. The proposed method, which is based on the relationship between blood pressure and the features of pulse waves, employs an inverse estimation and uses the blood pressure as the explanatory variable. Using this method, the blood pressure can be estimated with high accuracy even in situations where the pulse wave features are scattered, as the method uses the dynamic signal-to-noise ratio of the Taguchi method. In order to verify the effectiveness of the proposed method, we employed it to measure the systolic blood pressure. It could be confirmed that the estimation accuracy of the proposed method is higher than that of similar methods.

Heat capacity identification via Mahalanobis Taguchi System

This paper proposes a heat capacity identification method for cooking household appliances. In such appliances, a cooking flow is selected according to its cooking object capacity, hence identification of the heat capacity is a very important function. However, a conventional heat capacity identification method has been based on one variable using ldquoif-then rulesrdquo, thereby giving inaccurate results. This paper proposes a new heat capacity identification method that uses the well-known Mahalanobis-Taguchi system, which is similar to discriminant analysis. The effectiveness of this method is confirmed through an experiment.

Two-step variable screening method for the Mahalanobis-Taguchi method with small training data

This paper proposes a method that applies the Mahalanobis-Taguchi (MT) method in situations where the number of variables exceeds the number of training samples. In the proposed method, the diagnostic effectiveness of the variables is first evaluated using signal-to-noise (SN) ratio, and the MT method is then applied using only those variables with high diagnostic effectiveness. Accordingly, the variables are selected via a two-step process: (1) variables are selected before creating the unit space and (2) conventional variable selection is carried out using the MT method. The results of experimental evaluation of the effectiveness of the proposed method using four different datasets from the UCI Machine Learning Repository indicate that its diagnostic performance is superior to that of the RT method.