The Medical Information Database Network (MID-NET®) in Japan is a vast repository providing an essential pharmacovigilance tool. Gastrointestinal perforation (GIP) is a critical adverse drug event, yet no well-established GIP identification algorithm exists in MID-NET®. This study evaluated 12 identification algorithms by combining ICD-10 codes with GIP therapeutic procedures. Two sites contributed 200 inpatients with GIP-suggestive ICD-10 codes (100 inpatients each), while a third site contributed 165 inpatients with GIP-suggestive ICD-10 codes and antimicrobial prescriptions. The positive predictive values (PPVs) of the algorithms were determined, and the relative sensitivity (rSn) among the 165 inpatients at the third institution was evaluated. A trade-off between PPV and rSn was observed. For instance, ICD-10 code-based definitions yielded PPVs of 59.5

Drug-induced acute kidney injury (AKI) is a serious adverse drug reaction, which results in a significant decline in renal function and is known to progress to chronic kidney disease (CKD). Therefore, appropriate drug therapy is important to avoid the risk of drug-induced AKI and CKD, which are serious concerns in clinical practice. In this study, using the medical information database of Hamamatsu University Hospital, we investigated the risk factors that accelerate the onset of drug-induced AKI or its progression to CKD in patients who received aminoglycoside antibiotics (AGs) or glycopeptide antibiotics (GPs), which are strongly associated with drug-induced AKI and CKD. We performed logistic regression analysis using patients' background, laboratory test results, and concomitant drug use, among other such factors as explanatory variables and drug-induced AKI or CKD onset as objective variables to explore the risk factors for drug-induced AKI and CKD. Our results showed that co-administration of amphotericin B, piperacillin-tazobactam and other AGs and GPs, increased serum creatinine (Scr) and chloride concentrations, serum lactate dehydrogenase activity, and decreased serum albumin concentration were risk factors for drug-induced AKI onset. Moreover, a reduced blood urea nitrogen:Scr ratio at drug-induced AKI onset served as a risk factor for CKD. These results suggest that careful monitoring of the aforementioned factors is important to ensure appropriate usage of these drugs in patients treated with AGs and GPs.

Objectives In hypertrophic cardiomyopathy (HCM), specific ECG abnormalities are observed. Therefore, ECG is a valuable screening tool. Although several studies have reported on estimating the risk of developing fatal arrhythmias from ECG findings, the use of ECG to identify the severity of heart failure (HF) by applying deep learning (DL) methods has not been established.Methods We assessed whether data-driven machine-learning methods could effectively identify the severity of HF in patients with HCM. A residual neural network-based model was developed using 12-lead ECG data from 218 patients with HCM and 245 patients with non-HCM, categorised them into two (mild-to-moderate and severe) or three (mild, moderate and severe) severities of HF. These severities were defined according to the New York Heart Association functional class and levels of the N-terminal prohormone of brain natriuretic peptide. In addition, the patients were divided into groups according to Kansas City Cardiomyopathy Questionnaire (KCCQ)-12. A transfer learning method was applied to resolve the issue of the low number of target samples. The model was trained in advance using PTB-XL, which is an open ECG dataset.Results The model trained with our dataset achieved a weighted average F1 score of 0.745 and precision of 0.750 for the mild-to-moderate class samples. Similar results were obtained for grouping based on KCCQ-12. Through data analyses using the Guided Gradient Weighted-Class Activation Map and Integrated Gradients, QRS waves were intensively highlighted among true-positive mild-to-moderate class cases, while the highlighted part was highly variable among true-positive severe class cases.Conclusions We developed a model for classifying HF severity in patients with HCM using a deep neural network algorithm with 12-lead ECG data. Our findings suggest that applications of this DL algorithm for using 12-lead ECG data may be useful to classify the HF status in patients with HCM.

Deep learning technology has been used in the medical field to produce devices for clinical practice. Deep learning methods in cytology offer the potential to enhance cancer screening while also providing quantitative, objective, and highly reproducible testing. However, constructing high-accuracy deep learning models necessitates a significant amount of manually labeled data, which takes time. To address this issue, we used the Noisy Student Training technique to create a binary classification deep learning model for cervical cytology screening, which reduces the quantity of labeled data necessary. We used 140 whole-slide images from liquid-based cytology specimens, 50 of which were low-grade squamous intraepithelial lesions, 50 were high-grade squamous intraepithelial lesions, and 40 were negative samples. We extracted 56,996 images from the slides and then used them to train and test the model. We trained the EfficientNet using 2,600 manually labeled images to generate additional pseudo labels for the unlabeled data and then self-trained it within a student-teacher framework. Based on the presence or absence of abnormal cells, the created model was used to classify the images as normal or abnormal. The Grad-CAM approach was used to visualize the image components that contributed to the classification. The model achieved an area under the curve of 0.908, accuracy of 0.873, and F1-score of 0.833 with our test data. We also explored the optimal confidence threshold score and optimal augmentation approaches for low-magnification images. Our model efficiently classified normal and abnormal images at low magnification with high reliability, making it a promising screening tool for cervical cytology.

Purpose: To establish a new medical information database network (designated MID-NET (R)) to provide real-world data for drug safety assessments in Japan. Methods: This network was designed and developed by the Ministry of Health, Labour and Welfare and the Pharmaceuticals and Medical Devices Agency in collaboration with 23 hospitals from 10 healthcare organizations across Japan. MID-NET (R) is a distributed and closed network system that connects all collaborative organizations through a central data center. A wide variety of data are available for analyses, including clinical and administrative information. Several coding standards are used to standardize the data stored in MID-NET (R) to allow the integration of information originating from different hospitals. A rigorous and consistent quality management system was implemented to ensure that MID-NET (R) data are of high quality and meet Japanese regulatory standards (good post-marketing study practice and related guidelines). Results:MID-NET (R) was successfully established as a reliable and valuable medical information database and was officially launched in April 2018. High data quality with almost 100% consistency was confirmed between original data in hospitals and the data stored in MID-NET (R). A major advantage is that approximately 260 clinical laboratory test results are available for analysis. Conclusions: MID-NET (R) is expected to be a major data source for drug safety assessments in Japan. Experiences and best practices established in MID-NET (R) may provide a model for the future development of similar database networks.

Objectives: To summarize the activities of the International Academy of Health Sciences Informatics (IAHSI) in 2021 and welcome its 2021 Class of Fellows. Methods: Report on governance, strategic directions, newly elected fellows, plenary meetings, and other activities of the Academy. Results: As in 2020, all of the Academy's activities were carried out virtually due to the COVID-19 pandemic. In 2021, new Board members were elected. Strategic activities in data standards and interoperability and in mentorship moved forward. A new class of 26 Fellows was elected, bringing the total membership of the Academy to 204 Fellows from all regions of the world. In addition, a virtual plenary meeting was held. Conclusions: The Academy has continued to pursue its role as the honorific society globally for biomedical and health informatics. Expansion of strategic activities and membership will continue moving forward.

© 2022. IMIA and Thieme. This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/)

OBJECTIVES:To clarify the views of the general population of two countries (US and Japan), concerning handling of their medical records electronically, disclosure of the name of disease, secondary usage of information, compiling their records into a lifelong medical record, access to their medical records on the internet, questionnaire filling for delicate history, comprehensive consent for laboratory results, chart and genome profile, and AI use in diagnosis and explanation.METHODS:The authors contacted people nationwide in the United States at random via Random Digit Dialing (RDD) in 2008. Same questionnaire plus some new items were surveyed in 2022 by mail invited web entry. The authors had also surveyed people in Japan in 2007 and 2017 using same questionnaires sent by mail.RESULTS:In US, accessing own chart by internet became accepted (positive 52% to 61%) and popular in these 14 years. Japan showed small change, as regional medical record sharing is yet to come. About medical records in un-identifiable manner to be used for the purpose of medical error precautions, infectious disease measures and device/drug developments, in US, positive answers are constantly low, even for infectious disease prevention like CoVID-19. About preference to compile medical record into one file as a lifelong medical record, sharp contrast was observed. US people became favor of lifelong record (46% to 71%), while Japanese people decreased (76% to 57%). As for comprehensive consent, Japan positive answers are more than US for all situations, except if genome profile is included. US answers are almost same, even genome profile is included. About AI (artificial intelligence) application to healthcare, both US and Japan survey showed best preferred is "Doctor may use AI and everything, and explains in person". Japanese people largely prefer explanation in person, while US showed small preference.

This paper reports the history, background including politics, current status of Japan’s health imaging study and other information sharing. Its realization was slow until the Ministry of Health, Labour and Welfare (MHLW) started paying digital image storage at the same rate as films in 2008. Information sharing was initiated in early 2010s, which was before vendors became ready for Integrating the Healthcare Enterprise (IHE) cross-enterprise document sharing (XDS), with the result that most of 34 large regional sharing systems are in non-standardized protocol. One standardized example is the Hamamatsu area where inexpensive online PDI (portable data for imaging) was introduced.

This paper reports medical informatics conference held in November 2020, which is under influence of CoVID-19, in hybrid way. The methods taken to ensure presenters, chairpersons, participants, supporting staffs, vendors seamless infrastructure regardless of on line or on site. This made question answer session rich as usual. Some other outcomes newly made possible are reported, such as demonstration shown on big personal screen, much better than slide projector screen, new participants who became able to participate over prefecture boundaries in challenged times.

OBJECTIVES:To summarize the major activities of the International Academy of Health Sciences Informatics (IAHSI) in the 2020 time period and to welcome its 2020 Class of Fellows.METHOD:Report from the members of the Academy's Board.RESULTS:Due to the SARS-CoV-2 pandemic, both Plenary meetings in 2020 had to be organized as virtual meetings. Scientific discussions, focusing on mobilizing computable biomedical knowledge and on data standards and interoperability formed major parts of these meetings. A statement on the use of informatics in pandemic situations was elaborated and sent to the World Health Organization. A panel on data standards and interoperability started its work. 34 Fellows were welcomed in the 2020 Class of Fellows so that the Academy now consists of 179 members.CONCLUSIONS:There was a shift from supporting to strategic activities in the Academy's work. After having achieved organizational stability, the Academy can now focus on its strategic work and so on its main objective.

We studied the effectiveness of the direct data collection from electronic medical records (EMR) when it is used for monitoring adverse drug events and also detection of already known adverse events. In this study, medical claim data and SS-MIX2 standardized storage data were used to identify four diseases (diabetes, dyslipidemia, hyperthyroidism, and acute renal failure) and the validity of the outcome definitions was evaluated by calculating positive predictive values (PPV). The maximum positive predictive value (PPV) for diabetes based on medical claim data was 40.7% and that based on prescription data from SS-MIX2 Standardized Storage was 44.7%. The PPV for dyslipidemia was 50% or higher under either of the conditions. The PPV for hyperthyroidism based on disease name data alone was 20-30%, but exceeded 60% when prescription data was included in the evaluation. Acute renal failure was evaluated using information from medical records in addition to the data. The PPV for acute renal failure based on the data of disease names and laboratory examination results was slightly higher at 53.7% and increased to 80-90% when patients who previously had a high serum creatinine (Cre) level were excluded. When defining a disease, it is important to include the condition specific to the disease; furthermore, it is very useful if laboratory examination results are also included. Therefore, the inclusion of laboratory examination results in the definitions, as in the present study, was considered very useful for the analysis of multi-center SS-MIX2 standardized storage data.

BACKGROUND:Many countries adopt eHealth applications to support patient-centered care. Through information exchange, these eHealth applications may overcome institutional data silos and support holistic and ubiquitous (regional or national) information logistics. Available eHealth indicators mostly describe usage and acceptance of eHealth in a country. The eHealth indicators focusing on the cross-institutional availability of patient-related information for health care professionals, patients, and care givers are rare.OBJECTIVES:This study aims to present eHealth indicators on cross-institutional availability of relevant patient data for health care professionals, as well as for patients and their caregivers across 14 countries (Argentina, Australia, Austria, Finland, Germany, Hong Kong as a special administrative region of China, Israel, Japan, Jordan, Kenya, South Korea, Sweden, Turkey, and the United States) to compare our indicators and the resulting data for the examined countries with other eHealth benchmarks and to extend and explore changes to a comparable survey in 2017. We defined "availability of patient data" as the ability to access data in and to add data to the patient record in the respective country.METHODS:The invited experts from each of the 14 countries provided the indicator data for their country to reflect the situation on August 1, 2019, as date of reference. Overall, 60 items were aggregated to six eHealth indicators.RESULTS:Availability of patient-related information varies strongly by country. Health care professionals can access patients' most relevant cross-institutional health record data fully in only four countries. Patients and their caregivers can access their health record data fully in only two countries. Patients are able to fully add relevant data only in one country. Finland showed the best outcome of all eHealth indicators, followed by South Korea, Japan, and Sweden.CONCLUSION:Advancement in eHealth depends on contextual factors such as health care organization, national health politics, privacy laws, and health care financing. Improvements in eHealth indicators are thus often slow. However, our survey shows that some countries were able to improve on at least some indicators between 2017 and 2019. We anticipate further improvements in the future.

今回のAI（artificial intelligence）ブームは通算3回目のものであり，約35年前の前回のブームの際，著者も抗生剤選択支援システムを構築したが，実用にはならなかった．その理由は，データを入れる手間と，その推奨の根拠を示すことができなかったことによる．計算力の圧倒的進歩による今回のブーム到来は，判断するためのルールも自動で調整するという機能が実現している．膨大な候補化学構造とこれまた膨大なタンパク結合部位との組み合わせのうち，可能性の高そうなものを文献情報の意味解釈・シミュレーションによりリストアップすることで，創薬のコストはかなり改善する．しかし，診断結果の理由を問われると，生理学的・薬理学的根拠を示すことは未だ困難なようである．そもそも患者は本当のことを最初から話すであろうか．性感染症や精神疾患の過去の病歴を，それもそのデータは匿名化され，ビッグデータの一部になると言われて．著者が行ったアンケートでも，患者は人間の医師からの説明を好んでいることがわかった．インタビュー能力と説明能力こそは，なかなか機械化されないと考えられる一方で，ガイドラインを頭に叩き込んでそれを適用するという行為は，機械が得意な，取って代わられる可能性のあるものであろう．

Background: The Medical Information Database Network (MID-NET) is a national project that promotes effective safety measures for the active surveillance of drug safety assessments through pharmacoepidemiological methods, using real-world data in Japan. The MID-NET contains the data of approximately 5.05 million patients (as of December 2019) across 10 medical institutions, including 23 hospitals. One of the most important conditions for conducting pharmacoepidemiological research using multiple medical databases is to systematically verify of data standardization. Objectives: To evaluate the effect of improving the accuracy of standard data quality control by the development of a validation model for standard code mapping in multiple medical information databases. Methods: We established the standard code mapping validation center at one of the cooperating medical institutions of the MID-NET that could collect and manage information about the standard code interoperability. Additionally, we used the mapping table for the four standard codes, including the Japan Laboratory Test Standard Code, 10th Revision (JLAC10) code were collected from MID-NET cooperating institutions, and the accuracy of the mapping table was evaluated. Results: The observed four standard codes mapping ratio between institutions varied from >2,000 to <100. Moreover, the accuracies of standard codes were not standardized. We used a centralized standard code mapping validation model to provide feedback for standardizing JLAC-10 for each institution and meaningful differences between institutions were improved. Conclusions: The developed model visualized information differences and improved the data quality between multiple medical institutions.

Background: The International Academy of Health Sciences Informatics (IAHSI) is the Academy of the International Medical Informatics Association (IMIA). As an international forum for peers in biomedical and health informatics, the Academy shall play an important role in exchanging knowledge, providing education and training, and producing policy documents. Objectives: A major priority of the Academy’s activities in its inaugural phase was to define its strategy and focus areas in accordance with its objectives and to prioritize the Academy’s work, which can then be transferred to respective taskforces. Method: This document reflects the major outcomes of intensive discussions that occurred during 2019. It was presented at the Academy’s 3rd Plenary on August 25th, 2019, in Lyon, France. Results: Regardless of the ‘living nature’ of the strategy and focus areas document, it was concluded during the Plenary that the first version, which will be used as a base for decisions on the Academy’s future activities, should be made available to a broad audience. Three out of eight ‘Visions for IAHSI‘, presented in the IMIA Yearbook of Medical Informatics 2018, were identified as central for developing, implementing, and evaluating the Academy’s strategic directions: (1) advise governments and organizations on developing health and health sciences through informatics, (2) stimulate progress in biomedical and health informatics research, education, and practice, and (3) share and exchange knowledge. Taskforces shall be implemented to work in the following areas, which were considered as priority themes: (1) artificial intelligence in health: future collaboration of entities with natural and with artificial intelligence in health care, and (2) current landscape of standards for digital health. Conclusions: Taskforces are now being established. Besides specific key performance indicators, suggested for monitoring the work of theses taskforces, the strategy to monitor the progress of the Academy itself has to be measured by relevant and acceptable metrics.

Summary Objectives: To summarize the major activities of the International Academy of Health Sciences Informatics (IAHSI) from 2018 until 2019, and to provide an outline of actions planned for 2020. Method: Reporting about these activities and actions by the members of the Academy's first Board. Results: Academy bylaws were accepted by the Academy Plenary and the IMIA General Assembly on August 25th, 2019 and August 26th, 2019, respectively. Academy's strategy and focus areas were developed. Based on the Academy's eligibility criteria, the 2018/2019 Class made of 26 new Academy Fellows was elected. Future activities will concentrate on building taskforces for developing and implementing major Academy focus areas. Conclusions: We are glad to report that the Academy is strong and thriving.

It is expected that a large amount of data related to diabetes and other chronic diseases will be generated. However, databases constructed without standardized data item sets can be limited in their usefulness. To address this, the Collaborative Committee of Clinical Informatization in Diabetes Mellitus was established in 2011 by the Japan Diabetes Society and Japan Association for Medical Informatics. The committee has developed core item sets and self-management item sets for diabetes mellitus, hypertension, dyslipidemia, and chronic kidney disease in collaboration with the Japanese Society of Hypertension, Japan Atherosclerosis Society, Japanese Society of Nephrology, and Japanese Society of Laboratory Medicine, as well as a mapping table that aligns the self-management item sets with the Japanese standardized codes for laboratory testing. The committee also determined detailed specifications for implementing the four self-management item sets in personal health record applications to facilitate risk stratification, the generation of alerts using information and communications technology systems, the avoidance of data input errors, and the generation of reminders to input the self-management item set data. The approach developed by the committee may be useful for combining databases for various purposes (such as for clinical studies, patient education, and electronic medical record systems) and for facilitating collaboration between personal health record administrators.

What is known and objective This study used electronic medical records to identify risk factors and establish a detection algorithm for denosumab-induced hypocalcaemia. Methods We identified 201 patients with cancer who were initially prescribed denosumab. Hypocalcaemia was defined as an adjusted serum calcium level of <= 2.13 mmol/L. A diagnosis of denosumab-induced hypocalcaemia was confirmed by two physicians after reviewing patient medical records. We evaluated patient characteristics as potential screening factors. Moreover, a retrospective cohort study was conducted to identify risk factors for denosumab-induced hypocalcaemia. Odds ratios (ORs) were estimated using logistic regression analysis. Results We analysed 164 patients with a low risk of hypocalcaemia. Among these, 29 (17.7%) patients were suspected to have denosumab-induced hypocalcaemia. The times to onset of definitive hypocalcaemia were shorter among these patients than among patients with non-denosumab-induced hypocalcaemia. Based on receiver operating characteristic curve analysis, we used time to onset of hypocalcaemia of <= 90 days as a second screening factor. The positive predictive value of this factor was 87.5%. In the retrospective cohort study, a significant difference was observed among patients with serum alkaline phosphatase (ALP) levels of >5.95 mu kat/L before initial prescription (P < 0.01). Patients with higher serum ALP levels had a 6.63 times higher risk of developing hypocalcaemia than those without increased serum ALP levels (OR: 6.63, 95% confidence interval [CI]: 1.79-29.31). The same results were observed in a sensitivity analysis using another database. What is new and conclusion We developed a detection algorithm for denosumab-induced hypocalcaemia based on calcium levels and time to onset of hypocalcaemia. We also identified elevated ALP levels as a risk factor for hypocalcaemia. Clinicians should carefully monitor initial serum calcium levels and screen for signs of hypocalcaemia in patients receiving denosumab who demonstrate elevated serum ALP levels.