Improving the reliability of a physician's diagnosis also has a major impact on the patient. Studies related to liver diseases have been carried out since the early days of the Mahalanobis-Taguchi (MT) system, and in the MT methods that use inverse matrices, even when the true value is unknown, diagnostic decisions have been made with a standardized Mahalanobis distance scale by use of explanatory variables. When the true value is unknown, however, it is not easy to determine and study the discriminatory capability of the obtained distances or the suitability of their sensitivity. In the present study the correlation coefficients of the unit space were corrected by use of a discount coefficient method. The correction is shown to increase the distance values of the data and improve the S/N ratio of the discriminations. In the cases studied here, as compared with the conventional method, the detection sensitivity of the Mahalanobis distance for liver patients was improved by approximately a factor of two.
Genichi Taguchi described the role of quality engineering as freeing society from the need to predict, diagnose, and decide. In The Limits to Growth, the first report issued by the Club of Rome, Dennis Meadows et al. sounded the alarm by warning that if population growth and environmental destruction continued unabated, mankind's growth would reach its limit within 100 years due to depletion of resources and deterioration of the environment. In an initial report on research, inspired by the proposals of the Club of Rome, in which the Mahalanobis-Taguchi system was used to study changes in national census data, we showed that the distances of the error root mean square for each nation for each year displayed characteristic national traits with time-series trends. In this second report, the implications of the error root mean square in distinguishing unstable nations and developed nations from the features of each nation are studied in relation to itemized diagnostics by the MT system. The result is that, from the features of the gain of the percent of total population aged 65 or over' item, if positive and negative signs are assigned to the error root mean square distance, it is possible to distinguish between unstable nations and developed nations and use the MT system to understand the time-series changes for each nation.
The orthogonal arrays used to evaluate software are large because they are used for in-circuit tests after the software design phase. These evaluations take much time, and it is difficult to make exhaustive checks of the detailed specifications. Methods using small orthogonal arrays in the design process have been proposed and implemented, but because of a lack of understanding by designers and leaders, they have not been introduced into the design stage and fault detection has not been carried out during the design stage. The use of orthogonal arrays for software evaluation has taken hold, however, and it has been successfull in reducing defects. Practical proposals for introducing orthogonal arrays into the design process, and the challenges faced, are discussed in this report.