土井 俊祐

BACKGROUND:  When administering an infusion to a patient, it is necessary to verify that the infusion pump settings are in accordance with the injection orders provided by the physician. However, the infusion rate entered into the infusion pump by the health care provider cannot be automatically reconciled with the injection order information entered into the electronic medical records (EMRs). This is because of the difficulty in linking the infusion rate entered into the infusion pump by the health care provider with the injection order information entered into the EMRs. OBJECTIVES:  This study investigated a data linkage method for reconciling infusion pump settings with injection orders in the EMRs. METHODS:  We devised and implemented a mechanism to convert injection order information into the Health Level 7 Fast Healthcare Interoperability Resources (FHIR), a new health information exchange standard, and match it with an infusion pump management system in a standard and simple manner using a REpresentational State Transfer (REST) application programming interface (API). The injection order information was extracted from Standardized Structured Medical Record Information Exchange version 2 International Organization for Standardization/technical specification 24289:2021 and was converted to the FHIR format using a commercially supplied FHIR conversion module and our own mapping definition. Data were also sent to the infusion pump management system using the REST Web API. RESULTS:  Information necessary for injection implementation in hospital wards can be transferred to FHIR and linked. The infusion pump management system application screen allowed the confirmation that the two pieces of information matched, and it displayed an error message if they did not. CONCLUSION:  Using FHIR, the data linkage between EMRs and infusion pump management systems can be smoothly implemented. We plan to develop a new mechanism that contributes to medical safety through the actual implementation and verification of this matching system.

AIM: To investigate the impact of the COVID-19 pandemic and its preventive measures on the glycemic and lipid control in people with diabetes mellitus (DM). MATERIALS AND METHODS: We conducted this retrospective cohort study from April 2019 to March 2021; we termed the period from April 2019 to March 2020 as the pre-COVID-19 period, and the period from April 2020 to March 2021 as the COVID-19 period, and divided each of these two periods into four quarters. RESULTS: In the 1st quarter of the COVID period, when the Japanese government declared the first public health emergency, 3,465 people with diabetes mellitus were receiving treatment, which was 10.4% lower than that in the pre-COVID period. The annual mean HbA1c level was significantly elevated in the COVID-19 period. The annual mean total cholesterol (TC) and triglyceride (TG) levels were also significantly higher in the COVID-19 period. Although there were no significant differences in the glycemic control or annual medication between the two periods in people with type 1 diabetes mellitus, the annual mean HbA1c, TC, and TG levels were significantly higher in the COVID-19 period in people with type 2 diabetes mellitus. Furthermore, a significant increase in the percentage of prescriptions for glinides, biguanides, sodium-glucose cotransporter 2 inhibitors, and glucagon-like peptide-1 receptor agonists for people with type 2 diabetes mellitus was observed in the COVID period. CONCLUSIONS: It appears from our study that COVID-19 and its preventive measures had a negative impact on the glycemic and lipid control in people with diabetes mellitus.

AIMS/INTRODUCTION: To evaluate the impact of the COVID-19 pandemic on the glycemic control, eating habits, and body composition of people with diabetes mellitus; to identify the determinants of worsening glycemic control in people with diabetes mellitus. MATERIALS AND METHODS: This retrospective, longitudinal observational study was performed in outpatients with diabetes mellitus who visited our hospital between April 2019 and March 2020 (pre-COVID-19 period) and continued for follow up from April 2020 to March 2021 (COVID-19 period). We compared the glycemic control, nutritional intakes, and body composition of people with diabetes mellitus between the two periods. The changes in the HbA1c values (ΔHbA1c) and other study variables were compared between the two periods. Logistic regression analysis was performed to identify the factors associated with the increase of HbA1c levels. RESULTS: A significant increase of HbA1c was observed during the COVID-19 period. The percent fat mass (FM) also increased, while the percent skeletal muscle mass (SMM) decreased during the COVID-19 period. After adjustments for age and sex, the ΔBMI (OR:2.33), ΔFM (OR:1.45), and ΔSMM (OR:0.51) were identified as being associated with elevated levels of HbA1c. CONCLUSIONS: The COVID-19 pandemic had a negative impact on the glycemic control and body composition of people with diabetes mellitus. The increased body weight and FM and decreased SMM observed during the pandemic were associated with poor glycemic control in people with diabetes mellitus.

Background Delta check is widely used for detecting specimen mix-ups. Owing to the inadequate specificity and sparseness of the absolute incidence of mix-ups, the positive predictive value (PPV) of delta check is considerably low as it is labor consuming to identify true mix-up errors among a large number of false alerts. To overcome this problem, we developed a new accurate detection model through machine learning. Methods Inspired by delta check, we decided to conduct comparisons with the past examinations and broaden the time range. Fifteen common items were selected from complete blood cell counts and biochemical tests. We considered examinations in which ≥11 among the 15 items were measured simultaneously in our hospital; we created individual partial time-series data of the consecutive examinations with a sliding window size of 4. The last examinations of the partial time-series data were shuffled to generate artificial mix-up cases. After splitting the dataset into development and validation sets, we allowed a gradient-boosting-decision-tree (GBDT) model to learn using the development set to detect whether the last examination results of the partial time-series data were artificial mixed-up results. The model's performance was evaluated on the validation set. Results The area under the receiver operating characteristic curve (ROC AUC) of our model was 0.9983 (bootstrap confidence interval [bsCI]: 0.9983-0.9985). Conclusions The GBDT model was more effective in detecting specimen mix-up. The improved accuracy will enable more facilities to perform more efficient and centralized mix-up detection, leading to improved patient safety.

BACKGROUND: The uneven geographical distribution of physicians in Japan is a result of those physicians electing to work in certain locations. In order to understand this phenomenon, it is necessary to analyze the geographic movement of physicians across the Japanese landscape. METHODS: We obtained individual data on physicians from 1978 to 2012 detailing their attributes, work institutions, and locations. The data are from Japanese governmental sources (the Survey of Physicians, Dentists, and Pharmacists). The total sample size was 122 150 physicians, with 77.5% being male and 22.5% female. After obtaining the data, we calculated the geographical distance of each physician's movement by using geographic information systems software (GIS; ArcGIS, ESRI, Inc., CA, USA). Geographical distance was then converted into time distance. We compared the resulting median values through nonparametric testing and then conducted a multivariate analysis. Our next step involved the use of an age-period-cohort (APC) model to measure the degree of impact three points of data, experience (experience years), the historical and environmental context of the data (survey year), and physician cohort (registration year) had on the movement of each physician. RESULTS: The ratio of female physicians who selected an urban area as their first working location was higher than that of male physicians. However, the selection of an urban area was becoming more popular as a first working location for both males and females as the year of data increased. The overall distance of geographical movement for female physicians was less than it was for male physicians. Physicians moved the greatest distance between their second and fourth years following license acquisition, at which point the time distance became shorter. The median time distance was 46 min in 2000 and 22 min in 2008. The physicians in our study did not move far from their first working location, and the overall distance of movement lessened in the more recent years of study. The median distance of movement after 20 years was 25.9 km for male physicians, and 19.1 km for female physicians. The results of the APC model indicated that the effects of experience years (age) gradually declined, that the survey year (period) effects increased, and that the registration year (cohort) effects increased initially before leveling off. CONCLUSIONS: The trends following the introduction of the new mandatory training system in 2004 may imply that the concentration of physicians in Japan's urban areas is expected to increase. After 2000, the effect of that period on physicians explains their geographical movements more so than the factor of their age.

Accessibility to healthcare service providers, the quantity, and the quality of them are important for national health. In this study, we focused on geographic accessibility to estimate and evaluate future demand and supply of healthcare services. We constructed a simulation model called the patient access area model (PAAM), which simulates patients' access time to healthcare service institutions using a geographic information system (GIS). Using this model, to evaluate the balance of future healthcare services demand and supply in small areas, we estimated the number of inpatients every five years in each area and compared it with the number of hospital beds within a one-hour drive from each area. In an experiment with the Tokyo metropolitan area as a target area, when we assumed hospital bed availability to be 80%, it was predicted that over 78,000 inpatients would not receive inpatient care in 2030. However, this number would decrease if we lowered the rate of inpatient care by 10% and the average length of the hospital stay. Using this model, recommendations can be made regarding what action should be undertaken and by when to prevent a dramatic increase in healthcare demand. This method can help plan the geographical resource allocation in healthcare services for healthcare policy.

By the diffusion of Hospital Information Systems, many medical documents have been computerized. In addition, most of paper documents before computerization have been also scanned and archived as document images. These were usually converted to text data by using document analysis techniques and Optical Character Reader (OCR) and archived for medical document retrieval. However, the resolutions of some documents are not sufficient for character recognition because of storage spaces, scanning regulations and so on. Therefore, we cannot search desired keywords in the documents, as a result, these documents are not still used effectively in medical document retrieval systems. In this study, we discuss a keyword detection and extraction methods for these document images. As the first step of this study, this paper proposes a method to detect and extract desired words from these documents by using weighted dissimilarity and character sequence. Evaluation experiments using actual medical documents are conducted to discuss the effectiveness of the proposed method.

We performed the multi-year project to collect discharge summary from multiple hospitals and made the big text database to build a common document vector space, and developed various applications. We extracted 243,907 discharge summaries from seven hospitals. There was a difference in term structure and number of terms between the hospitals, however the differences by disease were similar. We built the vector space using TF-IDF method. We performed a cross-match analysis of DPC selection among seven hospitals. About 80% cases were correctly matched. The use of model data of other hospitals reduced selection rate to around 10%; however, integrated model data from all hospitals restored the selection rate.

We constructed a simulation model with a geographic information system (GIS) to predict the future shortage of beds in the Tokyo Metropolitan Area. With a grid square method, we calculated patient numbers for every 500 square meters of the Tokyo Metropolitan Area until 2040 and estimated whether those in need could be admitted to hospitals within an hour's drive from their homes. The simulation demonstrates that after 2025 many patients may not be able to find hospitals within this time framework. The situation will be especially serious in the center of Tokyo and along the railway lines, where many senior citizens reside. We can now apply this innovative GIS method in many fields and especially for the precise estimation of future demands for and supply of medical assistance.

In this study, we developed the Patient Access Area Model by using a Geographic information system (GIS), and, in order to evaluate the balance of medical supply and demand in the future in small areas, simulated patients' access to hospitals. We set the accessible area by patients' transit time for each hospital. The patients living in each 500 meters mesh were allowed to enter hospitals only within the access area. The hospitals have its limit to admit patients based on their actual numbers of beds. We distributed inpatients from each mesh across hospitals. For the evaluation of demand, if patients could not be distributed to the hospitals within the accessible area, we defined the situation as "over-demand." As a result, although it was expected that over 9000 inpatients will not receive inpatient care in a southwest area region in the studied prefecture, most of the over-demand is in the densely regions along large traffic lines in 2030. Using this model, we can know demand for local health resources more clearly. This method is very useful to plan geographical resource allocation in medical services. © 2013 The Authors.

We started a multi-year project to collect discharge summaries from multiple hospitals and create a big text database to build a common document vector space, and develop various applications such as the autoselection of the disease. As the first step, we extracted discharge summary from two hospitals. Using a text mining method, we carried out a DPC selection. There was a difference in term structure and number of terms between the discharge summaries from both hospitals. Nevertheless, the selection rate of the disease is resembled closely.

Chiba University Hospital website has a lot of information and helps visitors to find all kinds of information about the hospital, while too much information makes it difficult to find a suitable doctor as a specialist for them. Visitors have to search every department website one by one. This task is very troublesome, especially large hospital. To solve this problem, we developed a specialist search engine. By entering a disease name, the engine will find the specialists for the disease. The engine can control synonyms by using International Statistical Classification of Diseases and Related Health Problems (ICD), Japanese Standard Disease-Code Master and arrange the specialists in order of the experience. © 2012 IEEE.

Recently, a lot of paper-based documents used in hospitals have been computerized because of diffusion of Hospital Information Systems (HIS). However, some of previously scanned documents do not have enough resolution for document image processing, e. g. OCR Engine, due to storage limitation of the system. Currently, such documents are still archived in HIS, but not used effectively now. These should be converted to electrical data for resemble case search. This paper discusses document image processing methods to search low-resolution medical documents. As the first step of this study, we propose the tagging method for low-resolution document images archived in HIS. This paper shows the detail of the proposed method and experimental results to discusses the effectiveness of the proposed method. We also show problems and future works of this study in the end of paper.

Recently, Electronic Medical Record (EMR) systems have become popular in Japan, and numerous discharge summaries are being stored electronically, although they have not yet been reutilized. We performed text mining by using the term frequencyinverse document frequency method along with a morphological analysis of the discharge summaries from 3 hospitals (the Chiba University Hospital, St. Luke's International Hospital, and the Saga University Hospital). We found differences in the styles of the summaries between hospitals, while the rates of properly classified Diagnosis Procedure Combination (DPC) codes were almost the same. Beyond the different styles for the discharge summaries, the text mining method was able to obtain appropriate extracts of the proper DPC codes. An improvement was observed by using the integrated model data between the hospitals. It appeared that a large database containing data from many hospitals could improve the precision of text mining.

OBJECTIVES: The prevalence of electronic medical record in Japan varies according to the size of the hospital which is 62.5% in major hospitals, 21.7% in medium, 9.1% in small size hospitals, and 16.5% in clinics. The complete paperless system is very limited, though some major hospitals are aiming at this system. Several regional network systems which connect different platforms of EMRs, have been developing in many districts, while the final picture of a regional network has not been clearly proposed. To develop a whole electronic health record or personal health records system from the regional network data, we have several obstacles to overcome such as standardization, a privacy act, unique national health number. METHODS: Some experimental trials have just been started. The reuse of the accumulated data has also just been initiated. We exploited text mining systems (term frequency-inverse document frequency method) to find similar cases and auto-audit Japanese diagnosis related group (DRG) coding by using discharge summaries. RESULTS: The same or even a more extreme phenomenon of huge data accumulation is occurring in genetic research and confluence of multi-disciplines of informatics is the next step, which has an enormous accumulation of data and discoveries of the relations beyond the dimension of each informatics. CONCLUSIONS: We need another approach to science apart from the conventional method, and data-driven approach with data mining techniques must be brought in for each field. Informaticians have new important roles as coordinators to link up numerous phenomena over dimensions.

Recently, electronic medical record (EMR) systems have become popular in Japan, and number of discharge summaries is stored electronically, though they have not been reutilized yet. We performed text mining with Tf-idf method and morphological analysis in the discharge summaries from three Hospitals (Chiba University Hospital, St. Luke's International Hospital and Saga University Hospital). We showed differences in the styles of summaries, between hospitals, while the rate of properly classified DPC (Diagnosis Procedure Combination) codes were almost the same. Beyond different styles of the discharge summaries, text mining method could obtain proper extracts of proper DPC codes. Improvement was observed by using integrated model data between the hospitals. It seemed that huge database which contains the data of many hospitals can improve the precision of text mining.