Objective To investigate the current status and development of electronic health records (EHR) at home and abroad to grasp the development trends of EHR, so as to point out the direction of the development and relevant research on EHR. Methods Based on the Web of Science citation database and the principle of bibliometrics, we analyzed the retrieved literature in aspects of publication date, language, country/region, institution, author, etc. Results A total of 1 262 eligible studies were identified. The number of articles on EHR increased rapidly from only 2 in 1995 to 218 in 2012. In terms of country/region, the United States ranked the top in all countries (763 articles, accounting for 60.46%). In terms of institution, Harvard University ranked the top (135 articles, accounting for 10.70%). In terms of journal, the Journal of the American Medical Informatics Association ranked the top (106 articles, accounting for 8.40%). In terms of authors, David W. Bates ranked the top (45 articles, accounting for 3.57%). In terms of subject type, health care sciences services and medical informatics were mainly focused on. Conclusion The research on EHR has become a global hot spot and relevant bibliometrics will contribute to the timely and correctly grasp the whole picture of its development trends and main research direction.
Data integrity, accuracy, and traceability are key elements of high-quality clinical research, as well as weak links in the promotion of clinical research transparency. How to promote data quality has become a major concern to all clinical research stakeholders. In this article, we dissected and analyzed data generation and capturing process in clinical research, and identified a key aspect in improving data quality: to promote electronic source data, especially to break the barrier between electronic health records and clinical research systems. Additionally, we summarized the experiences regarding this issue in China and overseas to propose a solution suitable for China to improve data quality in clinical research: to strengthen clinical research source data management by building clinical research source data platform and adopt common source data management process in hospitals.
Currently, the development of deep learning-based multimodal learning is advancing rapidly, and is widely used in the field of artificial intelligence-generated content, such as image-text conversion and image-text generation. Electronic health records are digital information such as numbers, charts, and texts generated by medical staff using information systems in the process of medical activities. The multimodal fusion method of electronic health records based on deep learning can assist medical staff in the medical field to comprehensively analyze a large number of medical multimodal data generated in the process of diagnosis and treatment, thereby achieving accurate diagnosis and timely intervention for patients. In this article, we firstly introduce the methods and development trends of deep learning-based multimodal data fusion. Secondly, we summarize and compare the fusion of structured electronic medical records with other medical data such as images and texts, focusing on the clinical application types, sample sizes, and the fusion methods involved in the research. Through the analysis and summary of the literature, the deep learning methods for fusion of different medical modal data are as follows: first, selecting the appropriate pre-trained model according to the data modality for feature representation and post-fusion, and secondly, fusing based on the attention mechanism. Lastly, the difficulties encountered in multimodal medical data fusion and its developmental directions, including modeling methods, evaluation and application of models, are discussed. Through this review article, we expect to provide reference information for the establishment of models that can comprehensively utilize various modal medical data.