AMIA ... Annual Symposium proceedings. AMIA Symposium最新文献

Background: Lack of consensus on the appropriate look-back period for multi-drug resistance (MDR) complicates antimicrobial clinical decision support. We compared the predictive performance of different MDR look-back periods for five common MDR mechanisms (MRSA, VRE, ESBL, AmpC, CRE).

Methods: We mapped microbiological cultures to MDR mechanisms and labeled them at different look-back periods. We compared predictive performance for each look-back period-MDR combination using precision, recall, F1 scores, and odds ratios.

Results: Longer look-back periods resulted in lower odds ratios, lower precisions, higher recalls, and lower delta changes in precision and recall compared to shorter periods. We observed higher precision with more information available to clinicians.

Conclusion: A previously positive MDR culture may have significant enough precision depending on the mechanism of resistance and varying information available. One year is a clinically relevant and statistically sound look-back period for empiric antimicrobial decision-making at varying points of care for the studied population.

Search for information is now an integral part of healthcare. Searches are enabled by search engines whose objective is to efficiently retrieve the relevant information for the user query. When it comes to retrieving biomedical text and literature, Essie search engine developed at the National Library of Medicine (NLM) performs exceptionally well. However, Essie is a software system developed for NLM that has ceased development and support. On the other hand, Solr is a popular opensource enterprise search engine used by many of the world's largest internet sites, offering continuous developments and improvements along with the state-of-the-art features. In this paper, we present our approach to porting the key features of Essie and developing custom components to be used in Solr. We demonstrate the effectiveness of the added components on three benchmark biomedical datasets. The custom components may aid the community in improving search methods for biomedical text retrieval.

Intelligent prediction of risk of blood transfusion among hospitalized patients can identify at-risk patients and provide timely information to the hospital to plan and reserve resources to meet the demand of blood transfusion. While previously proposed solutions focus on sub-populations such as patients admitted to ICU after gastrointestinal bleeding or postpartum patients with hemorrhage, we design a predictive model applicable to complete in-patient population. Our model relies on patients' similarity graph based on temporal patterns among clinical history of the patients. These graphs are processed through graph convolutional neural network (GCNN) to estimate node or patient level risk of blood transfusion. Thus, our model not only learns from the patient's own clinical history but also from other patients with similar clinical history. The model is also capable of fusing diverse data elements from electronic health records (EHR) such as demographic information, billing codes, and recorded vital signs. Our model was validated on both internal and external sets and outperformed all comparative baseline models.

In December 2022, regulations from the U.S. Office of the National Coordinator for Health IT came into effect that require electronic health record (EHR) systems to accept the connection of any patient-facing digital health app using the SMART on FHIR standard. However, little has been reported with regard to architectural patterns that can be reused to take advantage of this industry development and integrate patient-facing apps into clinical workflows. To address this need, we propose SIMPLE, short for Standards-based Implementation Maximizing Portability Leveraging the EHR. The SIMPLE architectural pattern was designed to meet several key desiderata: do not require patients to install new software; do not retain patient data outside of the EHR; leverage EHRs' existing personal health record (PHR) capabilities to optimize user experience; and maximize portability. Using this pattern, an application for lung cancer screening known as MyLungHealth has been designed and is undergoing iterative user-centered enhancement.

More and more hospital Clinical Data Warehouses (CDWs) are developed to gain access to EHR data. The rapid growth of investments in CDWs suggest a real potential for innovation in healthcare. However, it is still not confirmed that CDWs will deliver on their promises as researchers working with CDWs face many challenges. To gain a better understanding of these challenges and how to overcome them, we conducted a series of semi-structured interviews with EHR data experts. In this article, we share some initial results from the ongoing interview study. Two main themes emerged from the analysis of the transcripts of the interviews: the importance of infrastructures in terms of data and how it is generated, and the difficulty to make care, clinical research, and data science work together. Finally, based on the experts' experience, several recommendations were identified when using a CDW.

The process of matching patients with suitable clinical trials is essential for advancing medical research and providing optimal care. However, current approaches face challenges such as data standardization, ethical considerations, and a lack of interoperability between Electronic Health Records (EHRs) and clinical trial criteria. In this paper, we explore the potential of large language models (LLMs) to address these challenges by leveraging their advanced natural language generation capabilities to improve compatibility between EHRs and clinical trial descriptions. We propose an innovative privacy-aware data augmentation approach for LLM-based patient-trial matching (LLM-PTM), which balances the benefits of LLMs while ensuring the security and confidentiality of sensitive patient data. Our experiments demonstrate a 7.32% average improvement in performance using the proposed LLM-PTM method, and the generalizability to new data is improved by 12.12%. Additionally, we present case studies to further illustrate the effectiveness of our approach and provide a deeper understanding of its underlying principles.

As Electronic Health Record (EHR) systems increase in usage, organizations struggle to maintain and categorize clinical documentation so it can be used for clinical care and research. While prior research has often employed natural language processing techniques to categorize free text documents, there are shortcomings relative to computational scalability and the lack of key metadata within notes' text. This study presents a framework that can allow institutions to map their notes to the LOINC document ontology using a Bag of Words approach. After preliminary manual value- set mapping, an automated pipeline that leverages key dimensions of metadata from structured EHR fields aligns the notes with the dimensions of the document ontology. This framework resulted in 73.4% coverage of EHR documents, while also mapping 132 million notes in less than 2 hours; an order of magnitude more efficient than NLP based methods.

This study explores the variability in nursing documentation patterns in acute care and ICU settings, focusing on vital signs and note documentation, and examines how these patterns vary across patients' hospital stays, documentation types, and comorbidities. In both acute care and critical care settings, there was significant variability in nursing documentation patterns across hospital stays, by documentation type, and by patients' comorbidities. The results suggest that nurses adapt their documentation practices in response to their patients' fluctuating needs and conditions, highlighting the need to facilitate more individualized care and tailored documentation practices. The implications of these findings can inform decisions on nursing workload management, clinical decision support tools, and EHR optimizations.

Uncertainty quantification in machine learning can provide powerful insight into a model's capabilities and enhance human trust in opaque models. Well-calibrated uncertainty quantification reveals a connection between high uncertainty and an increased likelihood of an incorrect classification. We hypothesize that if we are able to explain the model's uncertainty by generating rules that define subgroups of data with high and low levels of classification uncertainty, then those same rules will identify subgroups of data on which the model performs well and subgroups on which the model does not perform well. If true, then the utility of uncertainty quantification is not limited to understanding the certainty of individual predictions; it can also be used to provide a more global understanding of the model's understanding of patient subpopulations. We evaluate our proposed technique and hypotheses on deep neural networks and tree-based gradient boosting ensemble across benchmark and real-world medical datasets.

Although checklists can improve overall team performance during medical crises, non-compliant checklist use poses risks to patient safety. We examined how task attributes affected checklist compliance by studying the use of a digital checklist during trauma resuscitation. We first determined task attributes and checklist compliance behaviors for 3,131 resuscitation tasks. Using statistical analyses and qualitative video review, we then identified barriers to accurately tracking task status, finding that certain task attributes were associated with non-compliant checklist behaviors. For example, tasks with multiple steps were more likely to be incorrectly recorded as completed when the task was not performed to completion. We discuss challenges in capturing and tracking the status of tasks with attributes that contribute to non-compliant checklist use. We also contribute a framework for understanding how tasks with certain attributes can be designed on checklists to improve compliance.