A Systematic Review of Healthcare Information Technology Anomaly Classification
Explore the source record for details and available documents.
Pullum, Laura↗
Engineering topics
Drews, Frank
Publications and source records attributed to Drews, Frank.
Use of Event-Time Embeddings via RNN to Discern Novel Event Sequences in EHRs
In highly configurable health information technology (HIT) systems, such as VistA of the Veterans Health Administration, the variations in how the system is used among different healthcare facilities and how the data are recorded can be significant. Despite the successful standardization of care efforts, some of these variations can be indicative of HIT hazards and demand further investigation. In this work, we implemented a recurrent neural network (RNN) architecture to learn clinical provider order sequences and their temporal dynamics while predicting the orders' terminal state. We demonstrate model performance and provide a use case for the model discerning novel event sequences. This model is proposed to find novel event sequences in an operational environment.
Ozmen, Ozgur↗
A new methodological framework for hazard detection models in health information technology systems
Not Available
96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
Adaptive Anomaly Detection for Dynamic Clinical Event Sequences
Over the past decade, health information technology (IT) has enabled the amount of digital information stored in electronic health records (EHRs) to expand greatly. However, according to some studies, hazards in health IT can lead to changes in clinical decisions, care processes, and care outcomes, as well as other issues. Thus, the effects of health IT hazards on patient safety have been at the forefront of recent patient safety research. Nonetheless, hazard detection in health IT remains a challenge. In this paper, the authors assume that safety-related issues in health IT would exhibit anomalous characteristics in EHR data. Although all hazards will exhibit some anomalous characteristics, not all anomalies can be regarded as hazards. The authors hypothesize that errors in health IT could lead to interruptions in the sequence of clinical actions. To this end, the problem of detecting anomalous sequences in big EHR data is considered. This paper focuses on dynamic event sequences, which are a series of clinical actions in motion. The authors propose an adaptive anomaly detection approach that uses higher-order network representation to detect anomalous sequences. Furthermore, the authors propose a contiguous subsequence anomaly detection approach that identifies abnormal subsequences in the detected anomalous sequences. The proposed approaches are tested by using synthetic and real-world EHR data. The proposed methods outperform existing state of the art anomaly detection techniques. To reduce the computational complexity associated with the operational implementation of the proposed approaches, the Apache Spark environment was leveraged, and a much shorter run time together with improved performance were achieved, especially for data with more than 60,000 sequences.
Niu, Haoran↗
Detection of Anomalous Events in Electronic Health Records
Over the past decade, Health Information Technology (Health IT) has enabled an explosion in the amount of digital information stored in electronic health records (EHRs). According to recent studies, safety-related issues in healthcare can present themselves as anomalies in EHR data. Motivating examples of anomalous events in EHRs include clinical events related to invalid order cancellations or rejections, which may be initiated by clinical staff or automatic software routines in Health IT systems. Such events may be detected using anomaly detection or change point detection methods. In this paper, we explore the use of a forecasting approach to detect anomalies in EHR data using an online Support Vector Regression technique. Specifically, the proposed approach uses temporal frequency of activities in EHRs, coupled with dynamic robust confidence intervals, to characterize events as normal or anomalous. Once an event is characterized as an anomaly, our approach suppresses its effects in subsequent time intervals. The proposed approach shows encouraging results using real-world EHR data from the Veterans Affairs' corporate data warehouse
Pellett, Jordan J.↗
Anomaly Detection in Sequential Health Care Data using Higher-Order Network Representation
The ever-increasing use of information technology (IT) in health care presents new challenges to patient care. Information errors arising from the use of Health IT and their implications on care delivery and patient outcomes have been widely reported. Information errors can lead to changes in clinical decisions, care processes, and care outcomes, among others. An altered care process, for example, could interrupt the sequence of care, which could lead to changes in care decisions and/or changes in care outcomes. We define interruptions in the care process as anomalies in the care sequence. In this paper, we present a new approach based on the higher-order network (HON) representation to detect anomalies in sequential health care data using electronic health records. The results show that there are higher-order dependencies in health care data; and that the use of HON representation is more effective than the first-order network representation for detecting anomalies in sequential health care data.
Niu, Haoran↗
Feature Engineering and Process Mining to Enable Hazard Detection in Health Information Technology
Motivation for this work originates from a recognized need for developing data-driven approaches through analysis of complex electronic health records. Specifically, we aim to enhance the reliability of health information technology (HIT) systems by detection of plausible HIT hazards in clinical order transactions. In the absence of well-defined event logs in corporate data warehouses, our proposed approach identifies relevant timestamped data fields that could indicate transactions in the clinical order life cycle generating raw event sequences. Subsequently, we adopt state transitions of the OASIS Human Task standard to map the raw event sequences and simplify the complex process that clinical radiology orders go through. We describe how the current approach provides the potential to investigate areas of improvement and potential hazards in HIT systems. The discussion concludes with a use case and opportunities for future applications.
Ozmen, Ozgur↗
Method and apparatus for monitoring dynamic cardiovascular function using n-dimensional representatives of critical functions
A method, system, apparatus and device for the monitoring, diagnosis and evaluation of the state of a dynamic pulmonary system is disclosed. This method and system provides the processing means for receiving sensed and/or simulated data, converting such data into a displayable object format and displaying such objects in a manner such that the interrelationships between the respective variables can be correlated and identified by a user. This invention provides for the rapid cognitive grasp of the overall state of a pulmonary critical function with respect to a dynamic system.
Westinskow, Dwayne↗
Resource Management for Real-Time Adaptive Agents
Increased autonomy and automation in onboard flight systems offer numerous potential benefits, including cost reduction and greater flexibility. The existence of generic mechanisms for automation is critical for handling unanticipated science events and anomalies where limitations in traditional control software with fixed, predetermined algorithms can mean loss of science data and missed opportunities for observing important terrestrial events. We have developed such a mechanism by adding a Hierarchical Agent-based ReaLTime technology (HART) extension to our Dynamic Resource Management (DRM) middleware. Traditional DRM provides mechanisms to monitor the realtime performance of distributed applications and to move applications among processors to improve real-time performance. In the HART project we have designed and implemented a performance adaptation mechanism to improve reaktime performance. To use this mechanism, applications are developed that can run at various levels of quality. The DRM can choose a setting for the quality level of an application dynamically at run-time in order to manage satellite resource usage more effectively. A groundbased prototype of a satellite system that captures and processes images has also been developed as part of this project to be used as a benchmark for evaluating the resource management framework A significant enhancement of this generic mission-independent framework allows scientists to specify the utility, or "scientific benefit," of science observations under various conditions like cloud cover and compression method. The resource manager then uses these benefit tables to determine in redtime how to set the quality levels for applications to maximize overall system utility as defined by the scientists running the mission. We also show how maintenance functions llke health and safety data can be integrated into the utility framework. Once thls framework has been certified for missions and successfully flight tested it can be reused with little development overhead for other missions. In contrast, current space missions llke Swift manage similar types of resource trade -off completely with the scientific application code itself, and such code must be re-certified and tested for each mission even if a large portion of the code base is shared. This final report discusses some of the major issues motivating this research effort, provides a literature review of the related work, discusses the resource management framework and ground-based satellite system prototype that has been developed, indicates what work is yet to be performed, and provides a list of publications resulting from this work.
Welch, Lonnie↗
Obtaining the Greatest Scientific Benefit from Observational Platforms by Consideration of the Relative Benefit of Observations
One of the current trends in spacecraft software design is to increase the autonomy of onboard flight and science software. This is especially true when real-time observations may affect the observation schedule of a mission. For many science missions, such as those conducted by the Swift Burst Alert Telescope, the ability of the spacecraft to autonomously respond in real-time to unpredicted science events is crucial for mission success. We apply utility theory within resource management middleware to optimize the real-time performance of application software and achieve maximum system level benefit. We then explore how this methodology can be extended to manage both software and observational resources onboard a spacecraft to achieve the best possible observations.
Chelberg, David↗
An Optimization Framework for Dynamic, Distributed Real-Time Systems
Abstract. This paper presents a model that is useful for developing resource allocation algorithms for distributed real-time systems .that operate in dynamic environments. Interesting aspects of the model include dynamic environments, utility and service levels, which provide a means for graceful degradation in resource-constrained situations and support optimization of the allocation of resources. The paper also provides an allocation algorithm that illustrates how to use the model for producing feasible, optimal resource allocations.
Eckert, Klaus↗