In an intensive care unit, since a patient’s condition changes every moment, treatment-directions need to be changed rapidly according to it. The hospital information system which exists from before is an order entry system. Although it excelled in the goods claim and the medical-affairs claim, it was inadequate for precise medical-examination directions. On the other hand, creation of a thermometry table is a subject of department information system, and cooperation with an order-entry system was inadequate. The medical directions by the doctor and the directions receptacle by the nurse are required in order to achieve treatment procedure. Adequate goods supply and expense need are also important in the unit. The system for managing a medical treatment was defined as Clinical Management System (CMS). It managed directions, a directions receptacle, enforcement, and clinical record. The system for supporting CMS was defined as Hospital Administrating System (HAS). It managed the goods claim and the medical-affairs accounts claim. CMS creates claim information from clinical record after implementation according to directions and HAS arranges claim information to goods claim information and medical-affairs accounts claim information. In this system, a detailed directions change in intensive care unit was attained by managing medical directions and a claim order separately. Moreover, the stable goods supply and medical-affairs accounts claim was attained.

　In Kanazawa Medical University Hospital, the electronic medical record (EMR) system has been used to see patients from 2000. In order to promote cooperation between medical facilities, we constructed Kanazawa Medical University Oriented Regional Medical Information Network System (KORIN) to exhibit medical information of 150,000 patients which have been accumulated in our EMR system.

　We developed a medical information reference system in which a doctor in an outpatient clinic (hereafter clinic doctor) can refer to medical information generated in a core hospital of the area. A registered clinic doctor can refer to medical data concerning the only patient who was introduced to the hospital by the clinic doctor. We have already constructed a medical information database ‘EMIR; Essential Medical information Integrated Retrieval system.’ Requested data from a clinic doctor are taken from the EMIR and sent to a reference server which uses HL7 architecture. A clinic doctor can access the reference server via ISDN dial-up or on the Internet, and can refer to the patient’s data by common Web browser. To maintain the WWW’s original convenience and also to achieve security requirements, we used a single-sign on-tool which enables one-time password authentication without entering a password. We hope this system would contribute to smooth communication between a hospital and an outpatient clinic, and as well as to patient’s benefits.

　Group-based learning with presentation is now popular in Japan, because the course will contribute to the deep understanding of this field. However, it is very difficult to evaluate the group-based learning if this task aims not only at the evaluation of each group, but also at its contribution to learning of other groups. This paper proposes an approach to quantitative characterizations for students’ evaluation. For each presentation, evaluation form was distributed to tutors and students who were listening to the presentations of other groups. The collected evaluation forms were analyzed by rank correlation, corresponding analysis, cluster analysis (Ward’s method) and the product correlation method. The analysis results show the characteristics of students with respect to the presentations.

　Objectives: To study the ability of text mining technique for the selection of specific words related to diagnosis and to distinguish the diseases of discharge summaries. Materials and methods: 4,317 discharge summaries in Chiba University Hospital were selected out of 13 representative diseases. Diagnosis related terminological words were extracted by morphological analysis. Thus, the diseases were compared with each other using tf×idf vector space model and important specific words for each disease were selected. Furthermore, we applied the vector space model for new cases and indicated the vector by a radar chart. Results: 7,918 words were selected from cases and 74% of 390 cases were properly diagnosed. The maximum-tree problem and dendrogram method demonstrated reasonable relationships among 13 diseases. Conclusion: These results suggest the possibility that text-mining technique is applicable to the automotive classification of medical documents according to the diagnoses.

　Telemedicine using networked systems has recently come to be performed. Since various systems for the tele-surgery system environment such as a robot system, a simulation system, and a diagnostic system are individually developed and managed, cooperation between subsystems is often difficult to manage. This research aims at developing an application programming interface for a robot controller and improving the system employment to realize an organic interconnection of subsystems and an employment of subsystems sharing the same application programming interface.

　This paper describes the development of the real-time controller and the middleware of a robotic surgery system using CORBA (Common Object Request Broker Architecture), one of distributed object technologies. The validity of the developed system was examined with the verification experiment under various conditions applying to a remote surgical experiment. In the near future, the developed robot system and other subsystems, such as a visual information transmission system, are expected to be unified to construct an integrated operational environment for tele-surgery.

　We have proposed a new type of Hospital Information System (HIS) which can be utilized as an effective system of collecting and storing data in a suitable way for various kinds of medical process analyses. In this new system, data are extracted from HIS at the time of medical act order entry, and stored in the form of “medical process tag,” which has information terms of “Who”, “to Whom”, “Where”, “When”, “Why”, “What”, “How” (6W1H) with link information to detailed data. The term “Why” is linked to other 5W1H terms by selecting a purpose before making medical act order entry. Each medical process tag is stored as one record, and all medical process tags for all patients and all kinds of processes are stored in a single table. In this way, retrieval of necessary data for various analyses of medical processes involving many kinds of medical act orders from a large amount of data can be performed efficiently and flexible. “Medical process management system” based on this idea has been implemented as a part of new integrated medical information system at Kochi University Hospital since 2002.

　The purpose of this study was to develop tools to support admission schema and bed control for neurological patients. These actions have previously been difficult to adapt to clinical path. In this system, we defined medical care actions necessary for differential diagnoses as individual modules. For each medical care action, a weight on the basis of a time-variant constraint was assigned. When an attending physician chooses a differential diagnosis, our system lists the related medical care actions and calculates a weight on the basis of this time-variant constraint. When the attending physician executes a medical care action in this system, the system displays a patient list as an achievement rate calculated with each action's weight. This system is useful as a business support system for bed managers. In addition, we hope that we utilize this system as a database which can support the medical care of patients by attending physicians, as well as in case databases.

 D.P.C. system was enforced at Kobe University Hospital since July 2003. It was expected that this system made the doctor’s workload increase. For laborsaving of doctors, we decided to make the new D.P.C. coding system integrated with Diagnosis Coding System and Surgery Order Entry System. In these systems we had adopted ICD-10 and Japanese insurance coding system called K-Code, which were required in D.P.C. coding. When the doctors composed D.P.C. coding, they become possible to reuse former data about diagnosis and surgery operation, by using this system. In this paper, we discussed utility and a future vision of this system.

　Venders provide many kinds of medical information systems such as HIS, RIS and PACS. These commercially available systems are, however, expensive. Furthermore, detailed system architectures are, in many cases, not opened for users. To construct an integrated medical information system with the commercially available systems, we therefore have to spend much time and/or extra costs to link them each other. Consequently, we have been developed an integrated medical information system “KPECK,” that has been written primarily in c (Boland c) language, and distributed as open source codes. The KPECK consists of RIS and PACS and runs on the Linux OS. The system can be communicated with available HIS by TCP/IP sockets. The system has been worked without troubles for one year in a hospital with 340 beds. The strategy, disclosure of the system architecture as open source code, in the present project is benefit for us to further development of the system: many experts in the medical fields can be contributed to improve the system. Furthermore, users can be used KPECK “as is” without costs on software including development of the base of the system.

　We developed a Web based knowledge-base system accumulating essential information from studies of diagnostic accuracy with the function of searching and displaying those information. It is also available for performing systematic review among several reviewers. The accumulating information was comprised of bibliographic data, quality assessment and additional relevant data for adapting the evidence to clinical medicine as well as data of diagnostic performance. For performing systematic review, the system can display a table of quality assessment of primary studies in which any difference of it among each item validated by working group members is comparable, and a table of diagnostic accuracy of included studies with summary ROC curve and forest plots. Thus, it can provide a common collaborative workplace on the Internet for several reviewers. Moreover, this system is also available to register unpublished studies. Although it is still under development and it needs to overcome the several shortcomings such as security, we believe it has potential to be a useful knowledge-base system for the area of diagnostic test.

　Nurses looking after inpatients cannot take enough time for caring or managing patients because they have other time-consuming works, namely, keeping records of patients, establishing contact with doctors or nurses and so on. They usually memorize information about patients beside sickbed and write this information down on chart in nurse station. And so, paper based nursing record may trigger recording mistakes because the record depends on the memories of nurses. To address this issue, we have developed an electronic treatment planning system on bed-site. This is a specialized system for pressure ulcer using Personal Digital Assistant (PDA). Our system enables nurses to record, retrieve and change the patients’ record on bed-site directly over wireless LAN. In this paper, we describe how to design and implement the planning system, in particular how to design user interface and validity of PDA.

　Recently, pharmacists have been providing pharmaceutical management and counseling services. We have developed a supporting program for the services with personal digital assistants (PDA, Palm OS), in order to reduce laborious works for document preparation of medication instruction records. The records entered at the bedside, such as patient name, counseling date, compliance and medication instruction were transmitted from the database (HanDBase3.0, DDH Software) on PDA to the database (Access, Microsoft) on desktop PC by HotSync. This program has a high usability by using tabs, checkboxes, pop-ups. Our PDA program aims to support managing pharmaceutical counseling services at a point of act by pharmacists.

　Recently as the introduction of electronic medical recording system, the electronic processing of the complicated work in a hospital is progressing quickly. The introduction of electronic medical recording system will increase the efficiency of almost all hospital work, such as paper chart management, conveyance and various handwritten orders. On the other hand, it is pointed out that the increase in the efficiency will not be expected in the work field where a lot of deviation from the work process in the current electronic medical recording system exist or where the current electronic medical recording system does not assume to handle, i.e. the patient reception of an outpatient department, and that even the work load will increase in such occasions. The purpose of the present study is to derive the design of the effective supporting system which definitely enhances the efficiency of the labor by analyzing workers’ action and thinking process precisely in the work field where the current electronic medical recording system does not cover. As the first step of the purpose, we focused on the “patient reception work” at an outpatient department. We observed and analyzed the actions of the operating workers. As the result, we elucidated a lot of work processes including the implied rules and know-how which the patient receptionists used daily, and some important findings which have been overlooked when designing the support system for receptionist’s task.

　We have developed a regional network system for Chigasaki Medical Association with the aim of assisting information sharing and medical cooperation. The system also supports medical education after graduate for general practitioners. We made questionnaire surveys just before and six months after the beginning of practical use of the system in order to evaluate this system. Many responders including elderly associates considered that the BBS was the most valuable function of the system, and actually have used this function. Information sharing function for medical cooperation was not exciting to the users, but we expect that this function is instrumental for medical consumers in having more effective medical treatment. By using secured network system, such as VPN, we intend to make a refinement on the system in near future.

　Colors of real objects were hardly captured and reproduced using any present imaging system. Although color management technologies will compensate differences among a variety of imaging equipment, they do not necessarily reproduce exact colors of real objects.

　If spectral reflectance of a real object and spectral radiation of an illuminant are known, exact colors will be reproduced. However a color picture that includes spectral reflectance information for each pixel is hardly captured. Meanwhile, there is an emerging technology named ‘multispectral imaging,’ which will estimate reflectance spectra of skin and mucosa from a picture taken by a three-band digital camera. We applied this technology to develop an imaging system for absolute color appearance pictures, which will capture and reproduce exactly the same color as real lesion of skin and mucosa and succeeded in solution of major technological problems encountered.

　However, stable color reproduction has not been achieved by subjective tests at present. A great influence of chromatic adaptation of human eyes to color appearance of skin and mucosa was revealed through investigation of the cause of that instability.

　In medical information system, natural language processing interface will be useful if it realize. However, “Terminology Problem” arise. “Terminology Problem” have many aspects, such as standardization and lexicon harmonization. In order to find a solution to the problems, we focused on the systematization of terminological knowledge which is based on actual medical text “The Merck Manual.” In order to find out what kinds of terms are distinctively used in the medical knowledge architecture that The Merck Manual has, an automatic indexing method “TF/IDF weighting.” It was applied to perform statistically approximate calculation with the big medical terminology glossary which made by ours and thus, terms which depend of fields were extracted to develop a systematized lexicon.

　We can do more standardized, cost-effective, and qualified patient care by management with care pathways. Inpatient database during the recent period of 10 months was analyzed. The whole cases were classified into categories by diseases and the rate of use of care pathways was evaluated. An inpatient data sheet, including pathway data was made. Fifty-eight kinds of care pathways were used in our hospital (11 for gastroenterological disease, 11 for cardiac disease, 1 for neurological disease, 7 for renal disease, 11 for general surgery cases, 6 for cardiac surgery cases, and 11 for orthopedic surgery cases). Care pathways were used in 37% among 2,874 patients (26% of gastroenterological disease cases, 56% of cardiac disease cases, 28% of neurological disease, 10% of renal disease cases, 37% of general surgery cases, 41% of cardiac surgery cases, and 53% of orthopedic surgery cases). The inpatient data sheet was very useful to estimate the number of patients covered by newly developed care pathways in advance.

　In order to evaluate the students’ activeness and abilities of gathering and analyzing medical hot topics, questionnaires were made to the second and third grade medical students. In the questionnaire, students were asked on the knowledge and interest on severe acute respiratory syndrome (SARS). The students in the third grade have better knowledge on SARS than those in the second grade. This inaccurate knowledge of the disease of the second grade students was associated with the more serious impression to the disease. The students responded that general public had more awareness than themselves, and that they should have more awareness to the disease. These results suggest that in the medical education, we should promote the students’ interest with current medical topics and abilities of analyzing and conceptualizing medical information.

　It has been tried to clarify the functional characteristics of the Special Functioning Hospitals (SFH), based on their medical service area. We used the administrative data, which the Ministry of Health, Labour and Welfare required the SFHs to report in order to improve the DPC (Diagnosis Procedure Combination) system. Since this data format is common through all SFHs, it seems to be relatively easy to collect and utilize those data. In this research project we have a plan to widely collect data, collaborating with all SFHs in the Chugoku and Shikoku region. In this paper, we collected data as preliminary trial from the Okayama University Hospital and Kochi University Hospital. As a result of rough analysis, we could clarify the features of the medical service area and caring diseases for both SFHs.