Literature DB >> 11079884

New approaches to measuring the performance of programs that generate differential diagnoses using ROC curves and other metrics.

H S Fraser1, S Naimi, W J Long.   

Abstract

INTRODUCTION: Evaluation of computer programs which generate multiple diagnoses can be hampered by a lack of effective, well recognized performance metrics. We have developed a method to calculate mean sensitivity and specificity for multiple diagnoses and generate ROC curves.
METHODS: Data came from a clinical evaluation of the Heart Disease Program (HDP). Sensitivity, specificity, positive and negative predictive value (PPV, NPV) were calculated for each diagnosis type in the study. A weighted mean of overall sensitivity and specificity was derived and used to create an ROC curve. Alternative metrics Comprehensiveness and Relevance were calculated for each case and compared to the other measures.
RESULTS: Weighted mean sensitivity closely matched Comprehensiveness and mean PPV matched Relevance. Plotting the Physician's sensitivity and specificity on the ROC curve showed that their discrimination was similar to the HDP but sensitivity was significantly lower.
CONCLUSIONS: These metrics give a clear picture of a program's diagnostic performance and allow straightforward comparison between different programs and different studies.

Mesh:

Year:  2000        PMID: 11079884      PMCID: PMC2243809     

Source DB:  PubMed          Journal:  Proc AMIA Symp        ISSN: 1531-605X


  8 in total

1.  Development of a knowledge base for diagnostic reasoning in cardiology.

Authors:  W J Long; S Naimi; M G Criscitiello
Journal:  Comput Biomed Res       Date:  1992-06

2.  Differential diagnoses of the heart disease program have better sensitivity than resident physicians.

Authors:  H S Fraser; W J Long; S Naimi
Journal:  Proc AMIA Symp       Date:  1998

3.  Relationships among performance scores of four diagnostic decision support systems.

Authors:  E S Berner; J R Jackson; J Algina
Journal:  J Am Med Inform Assoc       Date:  1996 May-Jun       Impact factor: 4.497

4.  Enhancement of clinicians' diagnostic reasoning by computer-based consultation: a multisite study of 2 systems.

Authors:  C P Friedman; A S Elstein; F M Wolf; G C Murphy; T M Franz; P S Heckerling; P L Fine; T M Miller; V Abraham
Journal:  JAMA       Date:  1999-11-17       Impact factor: 56.272

5.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases.

Authors:  J A Hanley; B J McNeil
Journal:  Radiology       Date:  1983-09       Impact factor: 11.105

6.  The meaning and use of the area under a receiver operating characteristic (ROC) curve.

Authors:  J A Hanley; B J McNeil
Journal:  Radiology       Date:  1982-04       Impact factor: 11.105

7.  Performance of four computer-based diagnostic systems.

Authors:  E S Berner; G D Webster; A A Shugerman; J R Jackson; J Algina; A L Baker; E V Ball; C G Cobbs; V W Dennis; E P Frenkel
Journal:  N Engl J Med       Date:  1994-06-23       Impact factor: 91.245

8.  Validation of the AI/RHEUM knowledge base with data from consecutive rheumatological outpatients.

Authors:  H J Bernelot Moens
Journal:  Methods Inf Med       Date:  1992-09       Impact factor: 2.176

  8 in total
  1 in total

1.  Evaluation of a cardiac diagnostic program in a typical clinical setting.

Authors:  Hamish S F Fraser; William J Long; Shapur Naimi
Journal:  J Am Med Inform Assoc       Date:  2003-03-28       Impact factor: 4.497

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.