PubMed Health⌕ Search

Biomedical subjects

N L Jain

Publications and source records attributed to N L Jain.

13 recordsLinked to original sources
PubMed1997

Identification of findings suspicious for breast cancer based on natural language processing of mammogram reports.

There is need for encoded data for computerized clinical decision support, but most such data are unavailable as they are in free-text reports. Natural language processing offers one alternative for encoding such data. MedLEE is a natural language processing system which is in routine use for encoding chest radiograph and mammogram reports. In this paper, we study MedLEE's ability to identify mammogram findings suspicious for breast cancer by comparing MedLEE's encoding with a logbook of all suspicious findings maintained by the mammography center. While MedLEE was able to identify all the suspicious findings, it varied in the level of granularity, particularly about the location of the suspicious finding. Thus, natural language processing is a useful technique for encoding mammogram reports in order to detect suspicious findings.

Breast Neoplasms↗
PubMed1996

Evaluation of an objective plan-evaluation model in the three dimensional treatment of nonsmall cell lung cancer.

PURPOSE: Evaluation of three dimensional (3D) radiotherapy plans is difficult because it requires the review of vast amounts of data. Selecting the optimal plan from a set of competing plans involves making trade-offs among the doses delivered to the target volumes and normal tissues. The purpose of this study was to test an objective plan-evaluation model and evaluate its clinical usefulness in 3D treatment planning for nonsmall cell lung cancer. METHODS AND MATERIALS: Twenty patients with inoperable nonsmall cell lung cancer treated with definitive radiotherapy were studied using full 3D techniques for treatment design and implementation. For each patient, the evaluator (the treating radiation oncologist) initially ranked three plans using room-view dose-surface displays and dose-volume histograms, and identified the issues that needed to be improved. The three plans were then ranked by the objective plan-evaluation model. A figure of merit (FOM) was computed for each plan by combining the numerical score (utility in decision-theoretic terms) for each clinical issue. The utility was computed from a probability of occurrence of the issue and a physician-specific weight indicating its clinical relevance. The FOM was used to rank the competing plans for a patient, and the utility was used to identify issues that needed to be improved. These were compared with the initial evaluations of the physician and discrepancies were analyzed. The issues identified in the best treatment plan were then used to attempt further manual optimization of this plan. RESULTS: For the 20 patients (60 plans) in the study, the final plan ranking produced by the plan-evaluation model had an initial 73% agreement with the ranking provided by the evaluator. After discrepant cases were reviewed by the physician, the model was usually judged more objective or "correct." In most cases the model was also able to correctly identify the issues that needed improvement in each plan. Subsequent replanning confirmed that further manual plan optimization could be achieved in 17 patients. CONCLUSION: The objective plan-evaluation model was able to rank lung cancer radiotherapy plans from best to worst. It was useful in improving plans and may be useful to physicians in defining goals for patients based on the ability to effectively and safely treat their tumors.

Carcinoma, Non-Small-Cell Lung↗
PubMed1996

Identification of suspected tuberculosis patients based on natural language processing of chest radiograph reports.

Identification of eligible patients from electronically available patient data is a key difficulty in computerizing clinical practice guidelines because a large amount of the relevant data is stored as free text. We have been using MedLEE (Medical Language Extraction and Encoding System), a natural language processing system, to encode the clinical information in all chest radiograph and mammogram reports. This paper describes a retrospective study to determine if MedLEE can identify patients at risk for having tuberculosis (TB) based on their admission chest radiographs. Reports of 171 adult inpatients with culture-positive TB during 1992 and 1993 were manually coded (by a TB specialist) using seven terms suggestive of TB, and were also encoded by MedLEE. Using manual coding as the gold standard, MedLEE agreed on the classification of 152/171 (88.9%) reports--129/142 (90.8%) suspicious for TB and 23/29 (79.3%) not suspicious for TB; and 1072/1197 (89.6%) terms indicative of TB. Analysis showed that most of the discrepancies were caused by MedLEE not finding the location of the infiltrate. By ignoring the location of the infiltrate, the agreement became 157/171 (91.8%) reports and 946/1026 (92.2%) terms. Thus, natural language processing offers a practical alternative for using free-text reports to determine patient eligibility for computerized clinical practice guidelines.

Adult↗
PubMed1995

Using knowledge maintenance for preference assessment.

Most real-life decisions require the decision maker to make trade-offs in order to fulfill multiple conflicting objectives. This is especially true in medical decision making while selecting the optimal therapy plan from among competing therapy plans for a patient. Multi-attribute utility theory provides a framework to specify these trade-offs for optimal decision making based on the preferences of the decision maker. However traditional preference-assessment techniques are difficult to implement and rarely elicit the true preferences of the decision maker. We describe a new preference-assessment method based on the concept of knowledge maintenance where the preference model is changed each time it makes an incorrect recommendation. The method is implemented in a decision-theoretic system to evaluate competing three-dimensional radiation treatment plans. The preference-assessment method leads to preference models which perform better than preference models elicited using traditional assessment techniques.

Abdominal Neoplasms↗
PubMed1994

Integrated software tools for the evaluation of radiotherapy treatment plans.

PURPOSE: This article announces the availability of a convenient and useful software environment for the evaluation of three-dimensional (3D) radiotherapy treatment plans. MATERIALS AND METHODS: Using standards such as American National Standards for Information Systems C and the X Window System allowed us to bring the computation and display of dose-volume histograms, dose statistics, tumor control probabilities, normal tissue complication probabilities, and a figure of merit together under one user interface. These plan evaluation tools are not stand alone, but must interact with a 3D radiation therapy planning system to obtain the required dose matrices and patient anatomical contours. Installation of the software involves a programmer who writes a software bridge between the radiation therapy planning system and the tools, thereby providing access to local data files. This design strategy confines portability issues to one area of the software. RESULTS: Access to the other tools is through the Graphical Plan Evaluation Tool (GPET). GPET coordinates the use of each of the tools and provides graphical facilities for display of their results. Importantly, GPET assures that the displayed results of each tool have been computed with the same input specifications for all treatment plans being compared. For added convenience, the user can rearrange the resultant data to be reviewed in various ways on the video screen. The software design also allows incorporation of customized algorithms and input data for computing tumor control probability and normal tissue complication probabilities, since those currently available are controversial. CONCLUSION: The Graphical Plan Evaluation Tool unifies the simultaneous computation for several analytical tools and graphical display of their results. Within the constraints of the X Window System environment, this assemblage of software tools provides a portable, flexible, and convenient method for the quantitative evaluation of several radiotherapy treatment plans.

Dose-Response Relationship, Radiation↗
PubMed1994

Interactive electronic whiteboards in the medical classroom.

Most research on computer-assisted instruction has concentrated on developing systems to be used outside the teaching environment to supplement or complement in-class teaching. We believe that interactive large-screen computers can be used effectively in the classroom as electronic whiteboards to more effectively teach select medical school courses. We describe our experience with one such device, the Xerox LiveBoard, to teach a course on computer-assisted clinical decision analysis to a group of first-year medical students.

Computer-Assisted Instruction↗
PubMed1994

Construction of a literature database and its use to provide probabilities for decision-analytic models of thrombolytic therapy.

Probabilities for decision-analytic models are routinely obtained from the medical literature. This study describes development and use of a literature database to facilitate obtaining probabilities for decision-analytic models of thrombolytic therapy for acute myocardial infarction. Implementation demonstrates the concept of a literature database to be both feasible and effective. Specific difficulties encountered in the evaluation of continuous variables, the potential storage of actual probabilities, and the advantage for easy growth with the literature are discussed.

Databases, Bibliographic↗
PubMed1993

Objective evaluation of 3-D radiation treatment plans: a decision-analytic tool incorporating treatment preferences of radiation oncologists.

PURPOSE: Selecting the optimal radiation treatment plan from a set of competing plans involves making trade-offs among the doses delivered to the target volumes and normal tissues by the competing plans. Evaluation of 3-dimensional radiation treatment plans is difficult because it requires the review of vast amount of graphical and numerical data. We have developed an objective plan-ranking model based on the concepts of decision analysis. METHODS AND MATERIALS: Our model ranks a set of tentative radiation treatment plans from best to worst. A figure of merit is computed for each plan based on probabilities of possible clinical complications such as non-eradication of the tumor and radiation induced damage to the nearby healthy normal tissues, and weights which indicate their clinical relevance. This figure of merit is used to rank the plans. Key issues addressed by the model include the incorporation of individual treatment preferences of the radiation oncologist and clinical features of the patient. RESULTS: A methodology has been established for eliciting the treatment preferences of radiation oncologists. Results of this elicitation, and examples of several plan evaluations are presented. An interactive computer-based tool has been developed as one of a set of tools to assist in the evaluation of 3-dimensional radiation treatment plans. CONCLUSION: The paper presents a decision-analytic model incorporating radiation oncologists' treatment preferences and an interactive computer-based tool for objectively ranking competing radiation treatment plans. The tool can be used by radiation oncologists for the evaluation of competing plans, or as part of a system which tries to automatically generate optimal treatment plans using mathematical or symbolic techniques.

Decision Support Techniques↗
PubMed1993

Objective evaluation of radiation treatment plans.

The evaluation of radiation treatment plans involves making trade-offs among doses delivered to the tumor volumes and nearby normal tissues. Evaluating state-of-the-art three-dimensional (3D) plans is a difficult task because of the huge amount of planning data that needs to be deciphered. Multiattribute utility theory provides a methodology for specifying trade-offs and selecting the optimal plan from many competing plans. Using multiattribute utility theory, we are developing a clinically meaningful objective plan-evaluation model for 3D radiation treatment plans. Our model incorporates three of the factors involved in radiation treatment evaluation--treatment preferences of the radiation oncologist, clinical condition of the patient, and complexity of the treatment plan.

Decision Support Techniques↗
PubMed1992

Ranking radiotherapy treatment plans using decision-analytic and heuristic techniques.

Radiotherapy treatment optimization is done by generating a set of tentative treatment plans, evaluating them, and selecting the plan closest to achieving a set of conflicting treatment objectives. The evaluation of potential plans involves making trade-offs among competing possible outcomes. Multiattribute decision theory provides a framework for specifying such trade-offs and using them to select optimal actions. Using these concepts, we have developed a plan-ranking model which ranks a set of tentative treatment plans from best to worst. Heuristics are used to refine this model so that it reflects the clinical condition of the patient being treated and the practice preference of the physician prescribing the treatment. A figure of merit is computed for each tentative plan and is used to rank the plans. The approach described is very general and can be used for other medical domains having similar characteristics. The figure of merit can also be used as an objective function by computer programs that attempt to automatically generate an optimal treatment plan.

Decision Making, Computer-Assisted↗
PubMed1991

Ranking radiotherapy treatment plans using decision-analytic and heuristic techniques.

Radiotherapy treatment optimization is done by generating a set of tentative treatment plans, evaluating them and selecting the plan closest to achieving a set of conflicting treatment objectives. The evaluation of potential plans involves making tradeoffs among competing possible outcomes. Multiattribute decision theory provides a framework for specifying such tradeoffs and using them to select optimal actions. Using these concepts, we have developed a plan-ranking model which ranks a set of tentative treatment plans from best to worst. Heuristics are used to refine this model so that it reflects the clinical condition of the patient being treated and the practice preferences of the physician prescribing the treatment. A figure of merit is computed for each tentative plan, and is used to rank the plans. The approach described is very general and can be used for other medical domains having similar characteristics. The figure of merit can also be used as an objective function by computer programs that attempt to automatically generate an optimal treatment plan.

Artificial Intelligence↗
PubMedDate not supplied

Automated tuberculosis detection.

OBJECTIVE: To measure the accuracy of automated tuberculosis case detection. SETTING: An inner-city medical center. INTERVENTION: An electronic medical record and a clinical event monitor with a natural language processor were used to detect tuberculosis cases according to Centers for Disease Control criteria. MEASUREMENT: Cases identified by the automated system were compared to the local health department's tuberculosis registry, and positive predictive value and sensitivity were calculated. RESULTS: The best automated rule was based on tuberculosis cultures; it had a sensitivity of .89 (95% CI.75-.96) and a positive predictive value of .96 (.89-.99). All other rules had a positive predictive value less than .20. A rule based on chest radiographs had a sensitivity of .41 (.26-.57) and a positive predictive value of .03 (.02-.05), and rule the represented the overall Centers for Disease Control criteria had a sensitivity of .91 (.78-.97) and a positive predictive value of .15 (.12-.18). The culture-based rule was the most useful rule for automated case reporting to the health department, and the chest radiograph-based rule was the most useful rule for improving tuberculosis respiratory isolation compliance. CONCLUSIONS: Automated tuberculosis case detection is feasible and useful, although the predictive value of most of the clinical rules was low. The usefulness of an individual rule depends on the context in which it is used. The major challenge facing automated detection is the availability and accuracy of electronic clinical data.

Diagnosis, Computer-Assisted↗