There was no standard to evaluate the functions of electronic patient record(EPR) systems. This is a major obstacle to develop the “easy to use” EPR system. We developed “card type EPR system” by adding flexible interface layer to basic EPR system provided by the vender, that can change the input and output layouts and functions of EPR system, very easily as users need. We have been used this EPR system in out-patient consultation for three years. This paper considers the standard to evaluate the level of “easiness” to use the electronic patient record and compared “card type EPR system” and the basic EPR system. Both systems satisfy enough items in the general indicator, which is necessary to electronic patient records. Otherwise, in the 35 points relating to the user interface, 29 points satisfied in the “card type patient record system”, and 16 points in the basic system. We conclude that the “card type electronic patient record system” compensates the deficiency of the basic electronic patient record system and improves the usability.

View full abstract

 Data mining has been used in many fields, and recently these methods have drawn attention in the medical field as well. We compared 3 statistical methods with conventional linear discrimination analysis by hit ratio (prediction accuracy) to evaluate the best method in arteriosclerosis prediction. We enrolled 502 healthy people and diagnosed arteriosclerosis by measuring pulse wave velocity (baPWV), and constructed 4 models (linear discrimination, logistic regression, neural network and decision tree) to predict it. We compared precision once for internal check, and repeated the experiments 12 times for external check. The decision tree model had the best precision by both internal check (93.6%) and external check (83.2±2.9%). Though the linear discrimination model (79.8±9.3%) was inferior to the decision tree model (p value=0.002), it had a significantly higher hit ratio than the logistic regression model (75.4±2.7%) and the neural network (74.8±4.4%). The decision tree model showed the smallest standard deviation and the highest hit ratio in binary data prediction of arteriosclerosis. By using these techniques of multivariate analysis, in coordination with the electronic medical chart system, the automatic diagnostic system has the possibility of reducing human errors, such as misdiagnosis and underdiagnosis. This study indicated the decision tree (C 5.0 algorithm) model is one of the candidates to preliminary diagnose multifactorial diseases such as atherosclerosis.

View full abstract

 We have been operating a regional medical information sharing system, with the Munakata Medical Association Hospital as a central administrator, for a period of 2 years. The number of registered patients stands at 813 and the results during the period show medical institutions accessing information 1,839 times and 5,802 pages collated for reference.
 The analysis of the access log showed the following:
 1) Analysing the access log permits, evaluation of the information network.
 2) Reports and summarized information such as reports from medical image examinations, electronic patient records and document information were most used.
 3) Information on many of the patients who registered for the system was requested within the first two weeks of their registering, within six months covered 90% of the total accessed pages.
 4) Medical institutions accessed data in relation to two main categories of patients: those who were hospitalised or receiving continuous medical care; and those patients who had undergone workup.
 For a regional medical information sharing system, we cleared that it is more effective to computerize and utilize data where medical decisions and practice including reports and summaries are consolidated rather than simply inputting raw data such as medical examination data or medical images.

View full abstract