Automatic structuring of radiology free-text reports.

A natural language processor was developed that automatically structures the important medical information (eg, the existence, properties, location, and diagnostic interpretation of findings) contained in a radiology free-text document as a formal information model that can be interpreted by a computer program. The input to the system is a free-text report from a radiologic study. The system requires no reporting style changes on the part of the radiologist. Statistical and machine learning methods are used extensively throughout the system. A graphical user interface has been developed that allows the creation of hand-tagged training examples. Various aspects of the difficult problem of implementing an automated structured reporting system have been addressed, and the relevant technology is progressing well. Extensible Markup Language is emerging as the preferred syntactic standard for representing and distributing these structured reports within a clinical environment. Early successes hold out hope that similar statistically based models of language will allow deep understanding of textual reports. The success of these statistical methods will depend on the availability of large numbers of high-quality training examples for each radiologic subdomain. The acceptability of automated structured reporting systems will ultimately depend on the results of comprehensive evaluations.

Requirements of a Web-based experiment management system.

Recent advances in tools for scientific data acquisition, visualization, and analysis have lead to growing information management problems for medical research laboratories. An exponential increase in the volume of data, combined with a proliferation of heterogeneous formats and autonomous systems, has driven the need for flexible and powerful Experiment Management Systems (EMS). This paper provides a detailed analysis of the informatics requirements of an EMS, and proposes a new type of middleware called an EMS-Building Environment (EMSBE), which enables the rapid development of web-based systems for managing laboratory data and workflow. We describe the Web-Interfacing Respository Manager (WIRM), an open-source application server for building customizable experiment management systems. WIRM is being used to manage several ongoing experiments, including a natural language processor of radiological findings, and an interdisciplinary project for studying brain function.

A statistical natural language processor for medical reports.

Statistical natural language processors have been the focus of much research during the past decade. The main advantage of such an approach over grammatical rule-based approaches is its scalability to new domains. We present a statistical NLP for the domain of radiology and report on methods of knowledge acquisition, parsing, semantic interpretation, and evaluation. Preliminary performance data are given. A discussion of the perceived benefit, limitations and future work is presented.

Automated identification of episodes of asthma exacerbation for quality measurement in a computer-based medical record.

Harvard Community Health Plan and the Center for Intelligent Information Retrieval are developing tools to support automated quality fo care measurement from clinical text data. A statistically based text classification system uses semantic features in computerized encounter notes to identify acute exacerbations of asthma. Individual encounter notes are sorted in bins of highly likely, highly unlikely and uncertain likelihood of documenting exacerbation, and then aggregated into episodes of exacerbation for frequency analysis. It is estimated that this approach could reduce the burden of manual chart review by 65%.

Automated classification of encounter notes in a computer based medical record.

Harvard Community Health Plan is exploring emerging information technologies for means to use the text portion of its 25 year old computerized medical record system. The Center for Intelligent Information Retrieval is developing systems to answer the question: to what extent can automated information systems replace manual chart review of encounter notes? INQUERY, a probabilistic inference net information retrieval system, and FIGLEAF, an inductive decision tree text classifier are applied to the problem of classifying electronic encounter notes to identify acute exacerbations in pediatric asthmatics. Both systems achieve average precisions of greater than 80%, with a new enhancement to INQUERY's relevance feedback, the top performer. Refinement of the systems and plans for their integration are discussed.