Effective sparse imputation of patient conditions in electronic medical records for emergency risk predictions

Meta-clustering is a popular approach for finding multiple clusterings in the dataset, taking a large number of base clusterings as input for further user navigation and refinement. However, the effectiveness of meta-clustering is highly dependent on the distribution of the base clusterings and open challenges exist with regard to its stability and noise tolerance. In addition, the clustering views returned may not all be relevant, hence there is open challenge on how to rank those clustering views. In this paper we propose a simple and effective filtering algorithm that can be flexibly used in conjunction with any meta-clustering method. In addition, we propose an unsupervised method to rank the returned clustering views. We evaluate the framework (rFILTA) on both synthetic and real-world datasets, and see how its use can enhance the clustering view discovery for complex scenarios.     

ST-Audit: guideline-based automatic auditing of electronic patient records

This work presents the ST-Audit system that audits a patient record for conformance to a particular clinical guideline. The system uses ST-Guide which models a guideline as a set of states and transitions. The audit system tries to find a path in the state/transition diagram that corresponds to the actions taken by the physician, taking into consideration issues such as the unavailability of all the data needed to evaluate the transitions. The system was used to audit an outpatient clinic regarding their procedure for hypertension treatment (using the VI JNC guideline for hypertension), and the results of the number of non-compliant actions were presented and discussed. A follow up auditing showed a small but statistically significant reduction on the number of non-compliant actions for patients treated after the first audit.     

Human factors in patient safety as an innovation

The use of Human Factors and Ergonomics (HFE) tools, methods, concepts and theories has been advocated by many experts and organizations to improve patient safety. To facilitate and support the spread of HFE knowledge and skills in healthcare and patient safety, we propose to conceptualize HFE as innovations whose diffusion, dissemination, implementation and sustainability need to be understood and specified. Using Greenhalgh et al. (2004) model of innovation, we identified various factors that can either hinder or facilitate the spread of HFE innovations in healthcare organizations. Barriers include lack of systems thinking, complexity of HFE innovations and lack of understanding about the benefits of HFE innovations. Positive impact of HFE interventions on task performance and the presence of local champions can facilitate the adoption, implementation and sustainability of HFE innovations. This analysis concludes with a series of recommendations for HFE professionals, researchers and educators.     

中文电子病历中的时间关系识别

中文电子病历中的时间关系包括句内时间关系和句间时间关系,其中,句内时间关系包括句内事件-事件的时间关系和句内事件-时间的时间关系,句间时间关系即是句间事件-事件的时间关系。把中文电子病历文本中的时间关系识别转化成实体对分类问题,针对句内时间关系的识别,制定了高准确率的启发式规则,并设计了基本特征、短语句法特征、依存特征和其他特征,训练分类器缓解句内时间关系的识别错误;针对句间时间关系的识别,在高准确率的启发式规则之外,设计了基本特征、短语句法特征和其他特征,训练分类器减少句间时间关系的识别错误。实验结果表明,当分别使用支持向量机（SVM）、SVM和随机森林（RF）算法时,所提方法在句内事件-事件、句内事件-时间和句间事件-事件的时间关系识别上的效果最好,其F₁值分别达到了84.0%、85.6%和63.5%。     

Narrative electronic patient records as source of discharge diagnoses

OBJECTIVE: To evaluate the consistency of diagnostic data extracted from narrative electronic patient record (EPR) notes compared with the data from a patient administrative system (PAS). To assess potential benefit of using EPR notes as source of diagnosis data and as basis for case identification. DESIGN: Construction of a computer algorithm to extract ICD-9 codes from narrative EPR notes. Assessment of consistency and reliability of the diagnostic codes retrieved from EPR notes and PAS. Estimation of efficiency of case identification based on data from PAS and EPR. RESULTS: Diagnosis codes were retrieved from PAS with sensitivity of 0.989 and the positive predictive value (PPV) was 0.993. Codes were retrieved from EPR with sensitivity of 0.908 and PPV of 0.990. Combining these two sources increased sensitivity to 0.999. CONCLUSION: Discharge diagnoses were easily extracted from narrative EPR notes by automatic methods. Information extracted from record notes was not significantly different from the corresponding data in PAS, but EPR was incomplete as compared with PAS. Utilizing data extracted from EPR improved case identification significantly.     

Rapid application design of an electronic clinical skills portfolio for undergraduate medical students

The aim was to find how to use information and communication technology to present the clinical skills content of an undergraduate medical curriculum. Rapid application design was used to develop the product, and technical action research was used to evaluate the development process. A clinician-educator, two medical students, two computing science masters students, two other project workers, and a hospital education informatics lead, formed a design team. A sample of stakeholders took part in requirements planning workshops and continued to advise the team throughout the project. A university hospital had many features that favoured fast, inexpensive, and successful system development: a clearly defined and readily accessible user group; location of the development process close to end-users; fast, informal communication; leadership by highly motivated and senior end-users; devolved authority and lack of any rigidly imposed management structure; cooperation of clinicians because the project drew on their clinical expertise to achieve scholastic goals; a culture of learning and involvement of highly motivated students. A detailed specification was developed through storyboarding, use case diagramming, and evolutionary prototyping. A very usable working product was developed within weeks. "SkillsBase" is a database web application using Microsoft Active Server Pages, served from a Microsoft Windows 2000 Server operating system running Internet Information Server 5.0. Graphing functionality is provided by the KavaChart applet. It presents the skills curriculum, provides a password-protected portfolio function, and offers training materials. The curriculum can be presented in several different ways to help students reflect on their objectives and progress towards achieving them. The reflective portfolio function is entirely private to each student user and allows them to document their progress in attaining skills, as judged by self, peer and tutor assessment, and examinations. Training materials include web links and materials developed locally using pedagogic principles developed by the SkillsBase team. Although the usability of SkillsBase has been proven, uptake of software that has arisen 'bottom-up' from within the curriculum has proved slow. We plan to incorporate the SkillsBase services into a more comprehensive virtual managed learning environment, anticipating that presenting the functionality in an environment that is routinely used by students and teachers will increase uptake and use.     

Efficient and privacy-preserving similar electronic medical records query for large-scale ehealthcare systems

The advancements and adoption of cloud-assisted ehealthcare systems have enabled the storage of massive electronic medical records (EMRs) in the cloud for efficient and easy access. A direct benefit of EMRs is the ability of patients to search for EMRs that are similar to their own in the cloud for use as references. These similar EMRs can help a patient find appropriate medical services quickly. However, for large-scale ehealthcare systems, challenges remain with respect to ensuring the efficiency and privacy of these queries. In this study, we construct an efficient and privacy-preserving similar EMR query scheme to help patients find similar EMRs to reference in a large-scale ehealthcare system. Specifically, we propose a coarse-grained query method based on a binary decision tree to find a set of EMRs corresponding to the patient’s set of medical-symptom keywords. We also design a fine-grained query method to find similar EMRs that meet the threshold set by the patient. A detailed security analysis shows that the proposed scheme is secure. The efficiency of the proposed method in a large-scale ehealthcare system is verified experimentally.     

Improving the work efficiency of healthcare-associated infection surveillance using electronic medical records

In this study, we developed an integrated hospital-associated urinary tract infection (HAUTI) surveillance information system (called iHAUTISIS) based on existing electronic medical records (EMR) systems for improving the work efficiency of infection control professionals (ICPs) in a 730-bed, tertiary-care teaching hospital in Taiwan. The iHAUTISIS can automatically collect data relevant to HAUTI surveillance from the different EMR systems, and provides a visualization dashboard that helps ICPs make better surveillance plans and facilitates their surveillance work. In order to measure the system performance, we also created a generic model for comparing the ICPs’ work efficiency when using existing electronic culture-based surveillance information system (eCBSIS) and iHAUTISIS, respectively. This model can demonstrate a patient's state (unsuspected, suspected, and confirmed) and corresponding time spent on surveillance tasks performed by ICPs for the patient in that state. The study results showed that the iHAUTISIS performed better than the eCBSIS in terms of ICPs’ time cost. It reduced the time by 73.27 s, when using iHAUTISIS (114.26 s) and eCBSIS (187.53 s), for each patient on average. With increased adoption of EMR systems, the development of the integrated HAI surveillance information systems would be more and more cost-effective. Moreover, the iHAUTISIS adopted web-based technology that enables ICPs to online access patient's surveillance information using laptops or mobile devices. Therefore, our system can further facilitate the HAI surveillance and reduce ICPs’ surveillance workloads.     

浅谈数字认证技术在电子病历系统中的应用

电子病历,是近今年新生的产物,随着现代信息技术的发展,在医疗领域的应用,电子病历的出现大大提高了医务工作人员的工作效率,提升了医院管理水平,对医院的管理工作有着深远的影响.然而,在实施过程中,电子病历的真实性和安全性还应有效发挥数字认证技术的作用.本文通过分析电子病历系统应用中的安全漏洞问题,针对如何高质量的发挥数字认证技术在电子病历系统中的应用进行探讨,以期通过本文的阐述有效提升电子病历系统的安全性,加快医院信息化进程.     

基于XML的电子病历系统及其院际信息共享技术研究

随着现代信息技术的良好发展,医院信息化也呈现出较好的状态,在国内外的医疗领域当中关于电子病历的研究和开发都已经成为了重点关注的对象。在这其中,最为关键的就是要能够利用一种标准、开放且易于扩展的技术和方法来实现电子病历的功能。可扩展标记语言(XML)采取一种自我描述的方式来对数据结构进行定义,这样就能够实现在描述数据的同时描述结构,基于这样一种特点,XML技术在电子病历及其他一些医疗领域当中已经得到了广泛的研究和应用。本文正是在这样一种背景条件下,对电子病历的基本定义、XML的基本原理、XML在电子病历系统当中的应用以及院际信息共享技术等内容进行了分析和探讨,希望对这样一种技术的进一步探究和深入有所帮助。     

Are electronic medical records associated with improved perceptions of the quality of medical records, working conditions, or quality of working life?

The purposes of this study were to determine if users of electronic medical records (EMRs) perceived their medical records to be of higher quality than users of paper records and to examine the relationship between EMR use and perceptions of working conditions, quality of worklife and quality of care among family physicians. To do so, a cross-sectional survey of family physicians (n = 1482) was conducted. Survey items included measurement of use of an EMR, perceptions of medical records, working conditions, job satisfaction, and quality of care. One hundred and forty-three physicians (23%) reported using EMRs. Physicians who used EMRs were significantly more satisfied with the quality of their medical records. EMR use was not related to other outcomes. While EMR users value their record systems higher than non-users value the traditional system, EMR systems do not appear to directly impact the other variables. Indirect relationships are suspected and should be tested.     

RASSP: changing the paradigm of electronic system design

RASSP, a four-year US Department of Defense program, aims to reduce the cost and development time for military embedded digital signal processors by a factor of four. At the same time, RASSP will decrease signal processor life cycle costs by at least four times. This article summarizes the program's technical approach and provides measures, using real systems development examples, of progress made in the project's first two years     

Improving efficiencies and patient safety in healthcare through human factors and ergonomics

Efficiencies and patient safety can be improved through modeling of healthcare systems with consideration of human factors and ergonomics. With focus on predicting individual and system performance, safety and satisfaction can be considered from various perspectives including the clinician, patient, pharmacist and healthcare organization. The basis for predictions are models and hypotheses developed from scientific principles, methods and technology implementation strategies demonstrated, observed and reported through research. This paper provides a brief review and reappraisal of recent articles and book chapters to give some insights into how efficiencies and patient safety can be improved through human factors and ergonomics. Current needs and potential future research are also outlined. Web-resources from the World Health Care Congress, Agency for Healthcare Research & Quality, Institute for Healthcare Improvement, National Institutes of Health, the Veterans Administration, Regenstrief Center for Healthcare Engineering (RCHE) and the Institutional Review Board at Purdue, and recent special broadcasts from C-SPAN and PBS support and motivate further research. In considering the new research, it is proposed that a great deal more focus is needed on the process, methods and structuring of research studies, than on the implementations of a particular healthcare practice, technology or model. Students at cross-disciplinary research centers like RCHE at Purdue University are focusing on developing their ability to evaluate research based on the selected course readings and a “List of 10 ways” to look at a research paper systematically for critical analysis. With foundations in human factors and ergonomics it is expected that new potential healthcare systems engineering related research projects can be considered and developed through the semester projects and built upon as extensions that may lead to theses and dissertations. Recent project examples will be discussed. Research issues will be outlined along various health systems issues as follows: Modeling Systems and Economic Aspects, Healthcare Information Technologies, Electronic Prescribing and Computerized Physician Order Entry Systems, Cognitive Aspects and Human–Computer Interaction, Healthcare Work Analysis & Design, International and Cultural Issues, Human Factors, Ergonomics and Patient Safety, Technology Adoption, Data Collection and Analysis.     

Widening the scope of human factors safety assessment for decommissioning

In the nuclear industry, throughout the lifecycle of a nuclear plant or facility, operations vary in the degree of their tractability. Commissioned facilities should be relatively predictable in their operations, whereas decommissioning operations by their nature tend to be less predictable because they are a transition from one state to another, where safety systems, structures, components, and people serving safety may all change. Of course, different operations or states will have different characteristics and varying safety requirements. In this paper, we argue that by accounting for operations according to their relative tractability for their given phase in the lifecycle of a facility, a more targeted approach to managing safety is possible—one that recognises that the less tractable operation relies on human adaptation and resilience to a greater degree than the more tractable operation. From the outset, the safety of a decommissioning operation may therefore require a thorough account of the human and organisational factors that will support or impair human adaptation and resilience. High Reliability Organisation theory may provide a useful framework to consider how to engineer resilience in decommissioning and other operations of a less tractable nature.     

An organizational learning perspective on the assimilation of electronic medical records among small physician practices

Small physician practices play an essential role in the healthcare delivery system but are least likely to adopt health information technologies such as electronic medical records (EMRs). Factors contributing to low adoption include investment cost, productivity loss, and lack of financial incentives. However, these factors do not explain why some small practices, which face similar challenges nonetheless assimilate EMRs, while others do not. We investigated the assimilation of EMRs from the theoretic perspective of organizational learning in a survey of small physician practices and evaluated whether characteristics associated with organizational learning barriers are related to EMR assimilation. We found that learning-related scale, related knowledge, and diversity were positively associated with small physician practices' stage of assimilation of EMR technology. Our findings suggest that some small practices are able to overcome the substantial learning barriers presented by EMRs but that others will require support to develop sufficient learning capacity. We consider implications for practice from this research and areas requiring further research.