Statistical versus fuzzy measures of variable interaction in patients with stroke

Cathy M. Helgason, Davender S. Malik, Shih-Chuan Cheng, Thomas H. Jobe, John N. Mordeson

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Evidence-based medicine, founded in probability-based statistics, applies what is the case for the collective to the individual patient. An intuitive approach, however, would define structure in the (physiologic) system of interest, the human being, directly relevant to other systems (patients) composed of similar variables. A difference in measure of variable interaction in the patient from that in the collective would show how extrapolation of information from the latter to the single patient is counterintuitive. Methods: We compare statistical to 'fuzzy' measures of variable interaction. Three diagnostic variables are considered in 30 stroke patients who underwent the same diagnostic tests. 'Fit' (fuzzy information) values [0, 1] for degree of variable severity were expertly assigned by 2 blinded raters for real and fabricated patients. Fabricated patients were composed of real-patient 'fit' values after shuffling. Real and fabricated patients were each numerically represented as a set {blood, heart, vessel}. Three groups of fabricated patients and the real patient group were studied. Statistical [Pearson's product-moment (regression analysis) and Spearman's rank correlation] and three different fuzzy measures of variable interaction were applied to patient data. Results: Interaction for blood-vessel measured strong in real patients, and weak after one shuffle, using all fuzzy measures. By comparison, the same interaction was found in real patients by only 1 rater (Rater 2) using 1 statistical technique (Spearman's rank correlation) which, as did Pearson product-moment correlation, found a 'significant' interaction between blood-heart in fabricated patients. Conclusion: Our study suggests that the measure of variable interaction in nature - as combined in the individual (real) patient - is captured robustly by fuzzy measures and not so by standard statistical measures.

Original languageEnglish
Pages (from-to)77-84
Number of pages8
JournalNeuroepidemiology
Volume20
Issue number2
DOIs
StatePublished - 2001

Fingerprint

Stroke
Blood Vessels
Evidence-Based Medicine
Routine Diagnostic Tests
Regression Analysis

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Clinical Neurology

Cite this

Statistical versus fuzzy measures of variable interaction in patients with stroke. / Helgason, Cathy M.; Malik, Davender S.; Cheng, Shih-Chuan; Jobe, Thomas H.; Mordeson, John N.

In: Neuroepidemiology, Vol. 20, No. 2, 2001, p. 77-84.

Research output: Contribution to journalArticle

Helgason, Cathy M. ; Malik, Davender S. ; Cheng, Shih-Chuan ; Jobe, Thomas H. ; Mordeson, John N. / Statistical versus fuzzy measures of variable interaction in patients with stroke. In: Neuroepidemiology. 2001 ; Vol. 20, No. 2. pp. 77-84.
@article{566074814a0a45dba7339981c3784678,
title = "Statistical versus fuzzy measures of variable interaction in patients with stroke",
abstract = "Evidence-based medicine, founded in probability-based statistics, applies what is the case for the collective to the individual patient. An intuitive approach, however, would define structure in the (physiologic) system of interest, the human being, directly relevant to other systems (patients) composed of similar variables. A difference in measure of variable interaction in the patient from that in the collective would show how extrapolation of information from the latter to the single patient is counterintuitive. Methods: We compare statistical to 'fuzzy' measures of variable interaction. Three diagnostic variables are considered in 30 stroke patients who underwent the same diagnostic tests. 'Fit' (fuzzy information) values [0, 1] for degree of variable severity were expertly assigned by 2 blinded raters for real and fabricated patients. Fabricated patients were composed of real-patient 'fit' values after shuffling. Real and fabricated patients were each numerically represented as a set {blood, heart, vessel}. Three groups of fabricated patients and the real patient group were studied. Statistical [Pearson's product-moment (regression analysis) and Spearman's rank correlation] and three different fuzzy measures of variable interaction were applied to patient data. Results: Interaction for blood-vessel measured strong in real patients, and weak after one shuffle, using all fuzzy measures. By comparison, the same interaction was found in real patients by only 1 rater (Rater 2) using 1 statistical technique (Spearman's rank correlation) which, as did Pearson product-moment correlation, found a 'significant' interaction between blood-heart in fabricated patients. Conclusion: Our study suggests that the measure of variable interaction in nature - as combined in the individual (real) patient - is captured robustly by fuzzy measures and not so by standard statistical measures.",
author = "Helgason, {Cathy M.} and Malik, {Davender S.} and Shih-Chuan Cheng and Jobe, {Thomas H.} and Mordeson, {John N.}",
year = "2001",
doi = "10.1159/000054764",
language = "English",
volume = "20",
pages = "77--84",
journal = "Neuroepidemiology",
issn = "0251-5350",
publisher = "S. Karger AG",
number = "2",

}

TY - JOUR

T1 - Statistical versus fuzzy measures of variable interaction in patients with stroke

AU - Helgason, Cathy M.

AU - Malik, Davender S.

AU - Cheng, Shih-Chuan

AU - Jobe, Thomas H.

AU - Mordeson, John N.

PY - 2001

Y1 - 2001

N2 - Evidence-based medicine, founded in probability-based statistics, applies what is the case for the collective to the individual patient. An intuitive approach, however, would define structure in the (physiologic) system of interest, the human being, directly relevant to other systems (patients) composed of similar variables. A difference in measure of variable interaction in the patient from that in the collective would show how extrapolation of information from the latter to the single patient is counterintuitive. Methods: We compare statistical to 'fuzzy' measures of variable interaction. Three diagnostic variables are considered in 30 stroke patients who underwent the same diagnostic tests. 'Fit' (fuzzy information) values [0, 1] for degree of variable severity were expertly assigned by 2 blinded raters for real and fabricated patients. Fabricated patients were composed of real-patient 'fit' values after shuffling. Real and fabricated patients were each numerically represented as a set {blood, heart, vessel}. Three groups of fabricated patients and the real patient group were studied. Statistical [Pearson's product-moment (regression analysis) and Spearman's rank correlation] and three different fuzzy measures of variable interaction were applied to patient data. Results: Interaction for blood-vessel measured strong in real patients, and weak after one shuffle, using all fuzzy measures. By comparison, the same interaction was found in real patients by only 1 rater (Rater 2) using 1 statistical technique (Spearman's rank correlation) which, as did Pearson product-moment correlation, found a 'significant' interaction between blood-heart in fabricated patients. Conclusion: Our study suggests that the measure of variable interaction in nature - as combined in the individual (real) patient - is captured robustly by fuzzy measures and not so by standard statistical measures.

AB - Evidence-based medicine, founded in probability-based statistics, applies what is the case for the collective to the individual patient. An intuitive approach, however, would define structure in the (physiologic) system of interest, the human being, directly relevant to other systems (patients) composed of similar variables. A difference in measure of variable interaction in the patient from that in the collective would show how extrapolation of information from the latter to the single patient is counterintuitive. Methods: We compare statistical to 'fuzzy' measures of variable interaction. Three diagnostic variables are considered in 30 stroke patients who underwent the same diagnostic tests. 'Fit' (fuzzy information) values [0, 1] for degree of variable severity were expertly assigned by 2 blinded raters for real and fabricated patients. Fabricated patients were composed of real-patient 'fit' values after shuffling. Real and fabricated patients were each numerically represented as a set {blood, heart, vessel}. Three groups of fabricated patients and the real patient group were studied. Statistical [Pearson's product-moment (regression analysis) and Spearman's rank correlation] and three different fuzzy measures of variable interaction were applied to patient data. Results: Interaction for blood-vessel measured strong in real patients, and weak after one shuffle, using all fuzzy measures. By comparison, the same interaction was found in real patients by only 1 rater (Rater 2) using 1 statistical technique (Spearman's rank correlation) which, as did Pearson product-moment correlation, found a 'significant' interaction between blood-heart in fabricated patients. Conclusion: Our study suggests that the measure of variable interaction in nature - as combined in the individual (real) patient - is captured robustly by fuzzy measures and not so by standard statistical measures.

UR - http://www.scopus.com/inward/record.url?scp=0035001158&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035001158&partnerID=8YFLogxK

U2 - 10.1159/000054764

DO - 10.1159/000054764

M3 - Article

C2 - 11359073

AN - SCOPUS:0035001158

VL - 20

SP - 77

EP - 84

JO - Neuroepidemiology

JF - Neuroepidemiology

SN - 0251-5350

IS - 2

ER -