Browsing by Subject "information retrieval"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    A corpus-based approach for automated LOINC mapping
    (Oxford University Press, 2014-01-01) Fidahussein, Mustafa; Vreeman, Daniel J.; Department of Medicine, IU School of Medicine
    Objective To determine whether the knowledge contained in a rich corpus of local terms mapped to LOINC (Logical Observation Identifiers Names and Codes) could be leveraged to help map local terms from other institutions. Methods We developed two models to test our hypothesis. The first based on supervised machine learning was created using Apache's OpenNLP Maxent and the second based on information retrieval was created using Apache's Lucene. The models were validated by a random subsampling method that was repeated 20 times and that used 80/20 splits for training and testing, respectively. We also evaluated the performance of these models on all laboratory terms from three test institutions. Results For the 20 iterations used for validation of our 80/20 splits Maxent and Lucene ranked the correct LOINC code first for between 70.5% and 71.4% and between 63.7% and 65.0% of local terms, respectively. For all laboratory terms from the three test institutions Maxent ranked the correct LOINC code first for between 73.5% and 84.6% (mean 78.9%) of local terms, whereas Lucene's performance was between 66.5% and 76.6% (mean 71.9%). Using a cut-off score of 0.46 Maxent always ranked the correct LOINC code first for over 57% of local terms. Conclusions This study showed that a rich corpus of local terms mapped to LOINC contains collective knowledge that can help map terms from other institutions. Using freely available software tools, we developed a data-driven automated approach that operates on term descriptions from existing mappings in the corpus. Accurate and efficient automated mapping methods can help to accelerate adoption of vocabulary standards and promote widespread health information exchange.
  • Thumbnail Image
    Item
    Improving information retrieval from electronic health records using dynamic and multi-collaborative filtering
    (IEEE, 2019) Fan, Ziwei; Burgun, Evan; Ren, Zhiyun; Schleyer, Titus; Ning, Xia; Medicine, School of Medicine
    Due to the rapid growth of information available about individual patients, most physicians suffer from information overload when they review patient information in health information technology systems. In this manuscript, we present a novel hybrid dynamic and multi-collaborative filtering method to improve information retrieval from electronic health records. This method recommends relevant information from electronic health records for physicians during patient visits. It models information search dynamics using a Markov model. It also leverages the key idea of collaborative filtering, originating from Recommender Systems, to prioritize information based on various similarities among physicians, patients and information items We tested this new method using real electronic health record data from the Indiana Network for Patient Care. Our experimental results demonstrated that for 46.7% of test cases, this new method is able to correctly prioritize relevant information among top-5 recommendations that physicians are truly interested in.
  • Thumbnail Image
    Item
    Neurological Disorders and Publication Abstracts Follow Elements of Social Network Patterns when Indexed Using Ontology Tree-Based Key Term Search
    (Springer, 2014) Kulanthaivel, Anand; Light, Robert P.; Börner, Katy; Kong, Chin Hua; Jones, Josette F.; BioHealth Informatics, School of Informatics and Computing
    Disorders of the Central Nervous System (CNS) are worldwide causes of morbidity and mortality. In order to further investigate the nature of the CNS research, we generate from an initial reference a controlled vocabulary of CNS disorder-related terms and ontological tree structure for this vocabulary, and then apply the vocabulary in an analysis of the past ten years of abstracts (N = 10,488) from a major neuroscience journal. Using literal search methodology with our terminology tree, we find over 5,200 relationships between abstracts and clinical diagnostic topics. After generating a network graph of these document-topic relationships, we find that this network graph contains characteristics of document-author and other human social networks, including evidence of scale-free and power law-like node distributions. However, we also found qualitative evidence for Z-normal-type (albeit logarithmically skewed) distributions within disorder popularity. Lastly, we discuss potential consumer-centered as well as clinic-centered uses for our ontology and search methodology.