A practical method for transforming free-text eligibility criteria into computable criteria

Stanford Center for Biomedical Informatics Research, Stanford University, Stanford, CA 94305, USA.
Journal of Biomedical Informatics (Impact Factor: 2.48). 04/2011; 44(2):239-50. DOI: 10.1016/j.jbi.2010.09.007
Source: PubMed

ABSTRACT Formalizing eligibility criteria in a computer-interpretable language would facilitate eligibility determination for study subjects and the identification of studies on similar patient populations. Because such formalization is extremely labor intensive, we transform the problem from one of fully capturing the semantics of criteria directly in a formal expression language to one of annotating free-text criteria in a format called ERGO annotation. The annotation can be done manually, or it can be partially automated using natural-language processing techniques. We evaluated our approach in three ways. First, we assessed the extent to which ERGO annotations capture the semantics of 1000 eligibility criteria randomly drawn from Second, we demonstrated the practicality of the annotation process in a feasibility study. Finally, we demonstrate the computability of ERGO annotation by using it to (1) structure a library of eligibility criteria, (2) search for studies enrolling specified study populations, and (3) screen patients for potential eligibility for a study. We therefore demonstrate a new and practical method for incrementally capturing the semantics of free-text eligibility criteria into computable form.

Download full-text


Available from: Mor Peleg, Jul 07, 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Patients facing a serious disease often want to be able to search for relevant clinical trials for new or more effective alternative treatments. The NIH makes all of its trials available on a website, in fact, for this purpose. Its search facility, however, is difficult to use and requires the patient to sift through lengthy text descriptions for relevant information. Our overall aim is to build a system that allows for a more patient-focused clinical trial search facility. In this paper, we present a semantic integration approach using RDF triples to develop an integrated clinical trial knowledge representation, by linking different Linked Open Data such as clinical trials provided by NIH as well as the drug side effects dataset SIDER. The integration model uses UMLS to link concepts from different sources with consistent semantics and ontological knowledge. Patient-oriented functions that our prototype system provides include semantic search and query with reasoning ability, and semantic-link browsing where an exploration of one concept leads to other concepts easily via links which can provide visual search for the end users.
    Proceedings of the 12th IEEE Int. Conf. on Cognitive Informatics & Cognitive Computing, New York, NY; 07/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Clinical trials are conducted extensively in clinical research. Most often, the eligibility screening consists of manually re-viewing individual (patient) health records to identify po-tential candidates. Automated selection can improve the selection accuracy, increase the number of selected patients, and reduce the cost of the selection process. A formal rep-resentation of clinical trial eligibility criteria is however re-quired to facilitate such automated selection and also to en-able efficient multi-centric screening, thereby dealing with heterogeneous data coming from disparate sources. We present a novel framework to represent clinical trial eligibility criteria based on Topic Maps. The framework ad-dresses the needs in multi-centric screening by allowing for flexible and decentralized creation, sharing and integration of clinical trial representations and eligibility concepts. This is achieved through a formal but locally extensible ontology, thereby enabling unified querying across multiple institu-tions. Topic Maps is a semantic knowledge representation tech-nology for describing data and information. An ontology was created that describes the building blocks of a clinical trial. This representation encapsulates institution-specific data sources, thereby providing a domain-centric view on clinical trial definitions. Additionally, an evaluation engine was developed to assess patient eligibility using the ontol-ogy. A proof-of-concept was implemented at the Antwerp University Hospital. We show how the use of Topic Maps can facilitate patient recruitment across different centers.
    ACM SIGKDD Workshop on Health Informatics, HI-KDD 2012; 12/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The semi-automatic evaluation of eligibility crite-ria can facilitate the recruitment for clinical trials, timely com-pletion of studies and generation of clinical evidence about new approaches to treatment, prevention and diagnosis. Because eligibility criteria are represented as free text, automatically extracting their meaning and evaluating them for a particular patient is a challenging task. This paper presents our approach to the problem of automatic interpretation of criteria meaning. It is based on detecting in text semantic entities (diseases, treatment, measurements etc.) using ontology annotators and semantic taggers, and detecting predefined patterns provid-ing the contextual information in which these entities occur. Evaluation of the approach is the main subject of the paper. It covers several aspects: precision and recall of the pattern detection algorithm and the assessment of the implications of using the identified patterns to find potential candidates. It was performed manually using a subset of patterns and randomly selected 66 trials from The average precision and recall of pattern detection algorithm calculated for selected patterns is 0.91, meaning that in most cases using the patterns can lead to correct interpretation of criteria and can support patient recruitment.
    IEEE International Conference of Bioinformatics and Biomedicine, BIBM; 10/2012