“The effective utilization” of healthcare data and information is becoming increasingly important for the hospital management, the outcome evaluation, and decision-making to improve quality care and health services, as well as leaning and research, in Japanese medical institutions. Health information systems such as electronic medical records have been widely used, many organizations have massive amounts of data stored in various systems organization wide, then data warehouse (DWH) technologies are required to apply for information and evidence based managements.
 Although DWH technologies have been used in non-healthcare industries widely, but their used in healthcare industry is still in an early stage.
 In this symposium, speakers will present about their cases of DWH installations and the utilization of healthcare data/information in a single hospital or group hospitals. They also describe their benefits and issues according to their experiences. We will discuss about general and future aspects after the presentations.

View full abstract

 10 years will be passed in the next year after the ISO/TC215 was established to develop the international standard of health informatics. A lot of IS, TS and TR are published by ISO/TC215 and those activities are becoming to affect the development of HIS and EHR.
 In this progress, the relations with HL7 and CEN/TC251 those have the longer history of the standardization than ISO/TC215 will be the critical issues.
 The need of collaboration among HIT SDOs was recognized at the 1st Global HIT Standardization Summit which was held in Hamamatsu on September 2005. After the several meetings with ISO/TC215, HL7 and CEN/TC251, the Joint Initiative on SDO Global Health Informatics Standardization CHARTER was agreed by the 3 organizations at the end of August 2007.
 This paper reports the activities of ISO/TC215 and the harmonization plan with ISO/TC215, HL7 and CEN/TC215.

View full abstract

 1) Object: We have developed many kinds of medical and health information systems. These systems have huge databases now. But it was difficult for us to make evidences only with the data. We will discuss how we can make evidences with our experiences.
 2) Subjects and methods: We will discuss first how we can extract evidences from cross-sectional data with an example of results of health examination of about 700,000 persons. We will discuss secondly how we can make evidence from business application system with an example of health checkup system and resident registration system in cities.
 3) Results: We will show you we can get evidences from our information systems and some sampling method or follow-up systems. We have examined clinical guidelines with the evidences.
 4) Discussion: It is difficult to get any evidence only with data stored in information systems which we have developed. We can extract evidences with methods clearing vias and follow-up systems.

View full abstract

 Medical information flows through HIS/RIS/modalities/PACS, and each part has different specifications and limitations on handling Japanese character codes. The applicable combinations of Japanese character codes specified by standards such as DICOM or HL7 differ from sub-system to another, and this difference can cause problems in exchanging data. This paper shows some actual problems and solutions regarding this matter in the field. Also it proposes a standardized combination of Japanese character code sets for patient name, which is alphabets and 2-byte kanji characters.

View full abstract

 In our previous study, we applied frequency analysis on the sequential time study data of nurses in a busy in-patient ward, in order to extract periodic feature of the nursing service. In this paper, we prepared and analyzed a pair of simulated data of time study, in which either continuous or fragmented work schedule are assigned. Comparison of these simulated data with a pair of actual data, taken before and after the change of work schedule, was performed to examine whether the effect of the schedule change could be recognized and evaluated by the analysis. The simulated data was generated based on the total amount of classified items observed in the actual time study. The segment time series analysis was performed at the range of segment time series of 24hours and the shift of segment time of one hour for two days. The maximum entropy method (MEM) was used to calculate the power spectrum. The result of segment time series analysis of the simulated data featured a short cycle area and a medium to long cycle area. In the short cycle area, detailed variation of the cycle was clearly recognized, while the long term trend appeared in the medium to long cycle area. It is expected that the analysis of simulated data provides the visual assessment of the change of cycles of nursing service when applied to the data of actual time studies before and after the change of work schedule, as reported in our previous work. The simulation data analysis may be applied to evaluate the effect of other causes of interest.

View full abstract

 In recent days, a medical facility often collaborates the others in regions. There are several ways to collaborate medical facilities, thus it is important for medical facilities to understand how they stand. As a method of understanding their standpoints in regional medical collaboration, it is required that they analyze patients flow in a region. It is considered that we can use referral documents to analyze how the facility evaluates the others. In addition, it is also helpful to understand characteristics of the facility by analyzing them.
 In this study, we developed a system to collect referral documents’ data in one facility. Then we combined the referral documents’ data of inpatients with corresponding DPC (Diagnosis Procedure Combination) data and got the distribution of referral medical facilities for each MDC (Major Diagnosis Category) and each DPC disease category. In addition, we focused on cardiovascular diseases and got distribution of DPC disease category in the diseases. Finally, we analyzed how many referrals had come and where they had come from. As a result, we concluded the distribution of referral medical facilities can be varied by DPC disease category.

View full abstract

 We were aimed at clarifying it “how about organizational changes of metropolitan hospitals by introduction of electronic medical record systems” and “what kind of effect does it lead to” in this study. As a concrete method, we carried out questionary survey by Web for personnel and analyzed the result by statistical technique. Based on each factor extracted by factor analysis, we examined a factor score every hospital, relations between factors, a factor score every type of job. As a result, “a new function of electronic medical record systems” and “the result that country and Tokyo did not assume” were confirmed as factors, and a process a new function of electronic medical record systems passed through organizational changes, and to reach effects was shown. As for the medical record system, it is suggested that it is technology that an electronic medical record system needs an effort of the staff working at hospitals and organizational changes of hospitals.

View full abstract

 Hospital information system (HIS) with clinical decision support system (CDSS) is essential for providing safe and high quality care to patients. However, method how to incorporate CDSS into HIS is not fully established, and is still under consideration in most of medical institutions. In this paper, we describe about our data ware house (DWH) system for CDSS based on knowledge base technology, and discusses methodologies of integrating it with present HIS.

View full abstract

 We developed a MANACO (Metadata-based Analyzer for Network Appearances and Communications). It captures network packets on medical network directory with Layer1TAP device, and refers medical metadata based on ICD10. In this paper, we describe about a basic structure and implementation of our system, and a result of verification on real network environment with users.

View full abstract

 Home nursing care is in great need now because of the change of disease structure accompanying the aging, the shortening of the length of stay in hospital accompanying the revision of medical treatment fees, the introduction of care insurance since 2000 in care area, and the amendment of the insurance in 2005. However, supporting system for home nursing care does not provide the optimized knowledge to the individual and the symptom, and does provide generalized information only. Therefore, it is necessary to develop a system to provide an optimized coping strategy by relating personal information and concurrent disease, in addition to a basic data extracting coping strategy by estimating disease from symptom put into the computer. Therefore, we produced a database defined a relation of each element by elementizing and extracting the knowledge of home nursing care from existing literatures for 13 diseases developed by fever and the 10 concurrent disease, it aiming at using RDF. In addition, we produced a prototype for the system and considered to confirm if we can provide optimized knowledge from this database.

View full abstract

 Physicians’ inspection is usually made under various different illuminant conditions that must have significant influence on color of observing objects, and visual information coming into physicians’ eyes is always changing with time like a video image. Therefore, in electronic medical record, telemedicine and e-learning, reproduction of a video image with actual colors appearing under the observing illuminant condition would have significant advantage to reproduction of a still image with fixed colors. In visualizing spectral information unrecognizable by naked human eyes, real time processing would be much more effective for finding any morphological patterns common to a specific disease than providing elaborate still images. Based on these considerations, our system translates skin color captured by an ordinary three-band camera to that of an actual one regardless of the illuminating conditions and simultaneously visualizes distribution of melanin, hemoglobin and oxyhemoglobin in real time. The captured RGB raw image is converted into a spectral reflectance image by employing the Wiener estimation technique. Then the estimated reflectance spectra are converted into tristimulus values by using the spectral radiant distribution of the target illuminant and are also used to calculate the distribution of those pigments of the skin. In a preliminary experiment, professionals from a wide range of clinical fields pointed out various innovative uses of our system, which could not be brought out when the previous system that processes only still images was shown to them.

View full abstract