- Browse by Author
Browsing by Author "Biondich, Paul"
Now showing 1 - 10 of 11
Results Per Page
Sort Options
Item A Systematic Approach to Configuring MetaMap for Optimal Performance(Thieme, 2022) Jing, Xia; Indani, Akash; Hubig, Nina; Min, Hua; Gong, Yang; Cimino, James J.; Sittig, Dean F.; Rennert, Lior; Robinson, David; Biondich, Paul; Wright, Adam; Nøhr, Christian; Law, Timothy; Faxvaag, Arild; Gimbel, Ronald; Pediatrics, School of MedicineBackground: MetaMap is a valuable tool for processing biomedical texts to identify concepts. Although MetaMap is highly configurative, configuration decisions are not straightforward. Objective: To develop a systematic, data-driven methodology for configuring MetaMap for optimal performance. Methods: MetaMap, the word2vec model, and the phrase model were used to build a pipeline. For unsupervised training, the phrase and word2vec models used abstracts related to clinical decision support as input. During testing, MetaMap was configured with the default option, one behavior option, and two behavior options. For each configuration, cosine and soft cosine similarity scores between identified entities and gold-standard terms were computed for 40 annotated abstracts (422 sentences). The similarity scores were used to calculate and compare the overall percentages of exact matches, similar matches, and missing gold-standard terms among the abstracts for each configuration. The results were manually spot-checked. The precision, recall, and F-measure (β =1) were calculated. Results: The percentages of exact matches and missing gold-standard terms were 0.6-0.79 and 0.09-0.3 for one behavior option, and 0.56-0.8 and 0.09-0.3 for two behavior options, respectively. The percentages of exact matches and missing terms for soft cosine similarity scores exceeded those for cosine similarity scores. The average precision, recall, and F-measure were 0.59, 0.82, and 0.68 for exact matches, and 1.00, 0.53, and 0.69 for missing terms, respectively. Conclusion: We demonstrated a systematic approach that provides objective and accurate evidence guiding MetaMap configurations for optimizing performance. Combining objective evidence and the current practice of using principles, experience, and intuitions outperforms a single strategy in MetaMap configurations. Our methodology, reference codes, measurements, results, and workflow are valuable references for optimizing and configuring MetaMap.Item Do clinical decision-support reminders for medical providers improve isoniazid preventative therapy prescription rates among HIV-positive adults? Study protocol for a randomized controlled trial(BioMed Central, 2015-04-09) Green, Eric P.; Catalani, Caricia; Diero, Lameck; Carter, E. Jane; Gardner, Adrian; Ndwiga, Charity; Keny, Aggrey; Owiti, Philip; Israelski, Dennis; Biondich, Paul; Department of Medicine, IU School of MedicineBACKGROUND: This document describes a research protocol for a study designed to estimate the impact of implementing a reminder system for medical providers on the use of isoniazid preventative therapy (IPT) for adults living with HIV in western Kenya. People living with HIV have a 5% to 10% annual risk of developing active tuberculosis (TB) once infected with TB bacilli, compared to a 5% lifetime risk in HIV-negative people with latent TB infection. Moreover, people living with HIV have a 20-fold higher risk of dying from TB. A growing body of literature suggests that IPT reduces overall TB incidence and is therefore of considerable benefit to patients and the larger community. However, in 2009, of the estimated 33 million people living with HIV, only 1.7 million (5%) were screened for TB, and about 85,000 (0.2%) were offered IPT. METHODS/DESIGN: This study will examine the use of clinical decision-support reminders to improve rates of initiation of preventative treatment in a TB/HIV co-morbid population living in a TB endemic area. This will be a pragmatic, parallel-group, cluster-randomized superiority trial with a 1:1 allocation to treatment ratio. For the trial, 20 public medical facilities that use clinical summary sheets generated from an electronic medical records system will participate as clusters. All HIV-positive adult patients who complete an initial encounter at a study cluster and at least one return encounter during the study period will be included in the study cohort. The primary endpoint will be IPT prescription at 3 months post the initial encounter. We will conduct both individual-level and cluster-level analyses. Due to the nature of the intervention, the trial will not be blinded. This study will contribute to the growing evidence base for the use of electronic health interventions in low-resource settings to promote high-quality clinical care, health system optimization and positive patient outcomes. Trial registration ClinicalTrials.gov NCT01934309, registered 29 August 2013.Item Enhancing narrative clinical guidance with computer-readable artifacts: Authoring FHIR implementation guides based on WHO recommendations(Elsevier, 2021) Shivers, Jennifer; Amlung, Joseph; Ratanaprayul, Natschja; Rhodes, Bryn; Biondich, Paul; Herron School of Art and DesignIntroduction: Narrative clinical guidelines often contain assumptions, knowledge gaps, and ambiguities that make translation into an electronic computable format difficult. This can lead to divergence in electronic implementations, reducing the usefulness of collected data outside of that implementation setting. This work set out to evolve guidelines-based data dictionaries by mapping to HL7 Fast Health Interoperability Resources (FHIR) and semantic terminology, thus progressing toward machine-readable guidelines that define the minimum data set required to support family planning and sexually transmitted infections. Material and methods: The data dictionaries were first structured to facilitate mapping to FHIR and semantic terminologies, including ICD-10, SNOMED-CT, LOINC, and RxNorm. FHIR resources and codes were assigned to data dictionary terms. The data dictionary and mappings were used as inputs for a newly developed tool to generate FHIR implementation guides. Results: Implementation guides for core data requirements for family planning and sexually transmitted infections were created. These implementation guides display data dictionary content as FHIR resources and semantic terminology codes. Challenges included the use of a two-dimensional spreadsheet to facilitate mapping, the need to create FHIR profiles and resource extensions, and applying FHIR to a data dictionary that was created with a user interface in mind. Conclusions: Authoring FHIR implementation guides is a complex and evolving practice, and there are limited examples for this groundbreaking work. Moving toward machine-readable guidelines by mapping to FHIR and semantic terminologies requires a thorough understanding of the context and use of terminology, an applied information model, and other strategies for optimizing the creation and long-term management of implementation guides. Next steps for this work include validation and, eventually, real-world application. The process for creating the data dictionary and for generating implementation guides should also be improved to prepare for this expanding work.Item Envisioning health equity for American Indian/Alaska Natives: a unique HIT opportunity(Oxford, 2019) Cullen, Theresa; Flowers, Jan; Sequist, Thomas D.; Hays, Howard; Biondich, Paul; Laing, Maia Z.; Medicine, School of MedicineThe Indian Health Service provides care to remote and under-resourced communities in the United States. American Indian/Alaska Native patients have some of the highest morbidity and mortality among any ethnic group in the United States. Starting in the 1980s, the IHS implemented the Resource and Patient Management System health information technology (HIT) platform to improve efficiency and quality to address these disparities. The IHS is currently assessing the Resource and Patient Management System to ensure that changing health information needs are met. HIT assessments have traditionally focused on cost, reimbursement opportunities, infrastructure, required or desired functionality, and the ability to meet provider needs. Little information exists on frameworks that assess HIT legacy systems to determine solutions for an integrated rural healthcare system whose end goal is health equity. This search for a next-generation HIT solution for a historically underserved population presents a unique opportunity to envision and redefine HIT that supports health equity as its core mission.Item An Incremental Adoption Pathway for Developing Precision Medicine Based Healthcare Infrastructure for Underserved Settings(Medinfo 2017 Conference proceedings, 2017-08) Kasthurirathne, Suranga; Biondich, Paul; Mamlin, Burke; Cullen, Theresa; Grannis, ShaunRecent focus on Precision medicine (PM) has led to a flurry of research activities across the developed world. understaffed and underfunded health care systems in the US and elsewhere evolve to adapt PM to address pressing But how can healthcare needs? We offer guidance on a wide range of sources of healthcare data / knowledge sources as well as other infrastructure / tools that could inform PM initiatives, and may serve as low hanging fruit easily adapted on the incremental pathway towards a PM based healthcare system. Using these resources and tools, we propose an incremental adoption pathway to inform implementers working in underserved communities around the world on how they should position themselves to gradually embrace the concepts of PM with minimal interruption to existing care delivery.Item Information Discovery on Electronic Health Records Using Authority Flow Techniques(BMC, 2010-10-22) Hristidis, Vagelis; Varadarajan, Ramakrishna R.; Biondich, Paul; Weiner, Michael; Medicine, School of MedicineBackground As the use of electronic health records (EHRs) becomes more widespread, so does the need to search and provide effective information discovery within them. Querying by keyword has emerged as one of the most effective paradigms for searching. Most work in this area is based on traditional Information Retrieval (IR) techniques, where each document is compared individually against the query. We compare the effectiveness of two fundamentally different techniques for keyword search of EHRs. Methods We built two ranking systems. The traditional BM25 system exploits the EHRs' content without regard to association among entities within. The Clinical ObjectRank (CO) system exploits the entities' associations in EHRs using an authority-flow algorithm to discover the most relevant entities. BM25 and CO were deployed on an EHR dataset of the cardiovascular division of Miami Children's Hospital. Using sequences of keywords as queries, sensitivity and specificity were measured by two physicians for a set of 11 queries related to congenital cardiac disease.Results Our pilot evaluation showed that CO outperforms BM25 in terms of sensitivity (65% vs. 38%) by 71% on average, while maintaining the specificity (64% vs. 61%). The evaluation was done by two physicians. Conclusions Authority-flow techniques can greatly improve the detection of relevant information in EHRs and hence deserve further study.Item Keyphrase Identification Using Minimal Labeled Data with Hierarchical Context and Transfer Learning(medRxiv, 2023-05-26) Goli, Rohan; Hubig, Nina; Min, Hua; Gong, Yang; Sittig, Dean F.; Rennert, Lior; Robinson, David; Biondich, Paul; Wright, Adam; Nøhr, Christian; Law, Timothy; Faxvaag, Arild; Weaver, Aneesa; Gimbel, Ronald; Jing, Xia; Pediatrics, School of MedicineInteroperable clinical decision support system (CDSS) rules provide a pathway to interoperability, a well-recognized challenge in health information technology. Building an ontology facilitates creating interoperable CDSS rules, which can be achieved by identifying the keyphrases (KP) from the existing literature. However, KP identification for data labeling requires human expertise, consensus, and contextual understanding. This paper aims to present a semi-supervised KP identification framework using minimal labeled data based on hierarchical attention over the documents and domain adaptation. Our method outperforms the prior neural architectures by learning through synthetic labels for initial training, document-level contextual learning, language modeling, and fine-tuning with limited gold standard label data. To the best of our knowledge, this is the first functional framework for the CDSS sub-domain to identify KPs, which is trained on limited labeled data. It contributes to the general natural language processing (NLP) architectures in areas such as clinical NLP, where manual data labeling is challenging, and light-weighted deep learning models play a role in real-time KP identification as a complementary approach to human experts' effort.Item OpenMRS, A Global Medical Records System Collaborative: Factors Influencing Successful Implementation(2011-10) Mohammed-Rajput, Nareesa A.; Smith, Dawn C.; Mamlin, Burke; Biondich, Paul; Doebbeling, Bradley N.OpenMRS is an open-source, robust electronic health record (EHR) platform that is supported by a large global network and used in over forty countries. We explored what factors lead to successful implementation of OpenMRS in resource constrained settings. Data sources included in-person and telephone key informant interviews, focus groups and responses to an electronic survey from 10 sites in 7 countries. Qualitative data was coded through independent coding, discussion and consensus. The most common perceived benefits of implementation were for providing clinical care, reporting to funders, managing operations and research. Successful implementation factors include securing adequate infrastructure, and sociotechnical system factors, particularly adequate staffing, computers, and ability to use software. Strategic and tactical planning were successful strategies, including understanding and addressing the infrastructure and human costs involved, training or hiring personnel technically capable of modifying the software and integrating it into the daily work flow to meet clinicians’ needs.Item Towards Standardized Patient Data Exchange: Integrating a FHIR Based API for the Open Medical Record System(IOS, 2015) Kasthurirathne, Suranga Nath; Mamlin, Burke; Grieve, Grahame; Biondich, Paul; Department of BioHealth Informatics, School of Informatics and ComputingInteroperability is essential to address limitations caused by the ad hoc implementation of clinical information systems and the distributed nature of modern medical care. The HL7 V2 and V3 standards have played a significant role in ensuring interoperability for healthcare. FHIR is a next generation standard created to address fundamental limitations in HL7 V2 and V3. FHIR is particularly relevant to OpenMRS, an Open Source Medical Record System widely used across emerging economies. FHIR has the potential to allow OpenMRS to move away from a bespoke, application specific API to a standards based API. We describe efforts to design and implement a FHIR based API for the OpenMRS platform. Lessons learned from this effort were used to define long term plans to transition from the legacy OpenMRS API to a FHIR based API that greatly reduces the learning curve for developers and helps enhance adhernce to standards.Item Trends in RED Blood CELL Transfusions within the Biologics Effectiveness and Safety (BEST) Initiative Network, 2012-2018(AABB, 2019-09) Obidi, Joyce; Villa, Carlos; Storch, Emily; Whitaker, Barbee; Chada, Kinnera; Williams, Alan; Fowler, Stephanie; Schilling, Lisa M.; Kahn, Michael G.; Edlavitch, Stanley A.; Purkayastha, Saptarshi; Hayden, Ross; Biondich, Paul; Falconer, Thomas; Natarajan, Karthik; Reich, Christian; Shoaibi, AzadehThe U.S. FDA Center for Biologics Evaluation and Research (CBER) recently established the Biologics Effectiveness and Safety (BEST) Initiative which is an active surveillance system for all CBER regulated products, including blood components, and which also allows for investigation of patterns in utilization. The 2015 National Blood Collection& Utilization Survey (NBCUS) Report found an overall decrease in the number of units of red blood cells (RBCs) transfused in recent years. The aim of this study is to describe the utilization of RBCs from 2012 through 2018 using electronic health records (EHR) from three data sources participating in the BEST Initiative.