Medical pattern recognition pitfalls in a clinical setting: Renal cell carcinoma survival prediction

C. Kimme-Smith, S. T. Cochran, Denise Roe

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Three studies to predict renal cell carcinoma patient survival from tumor textures found in digitized early venous phase arteriograms were successful individually. However, when the three methods were compared, they were not consistent and no single method was clinically useful. The first study predicted 5 yr survival of 37 patients with 87% accuracy. The second study added 29 patients to the data base; the poor survival of the 21 patients who died within 5 yr of diagnosis was predicted with 80.9% accuracy. When 27 of these cases were redigitized with a laser scanner, average survival prediction accuracy was 78%. In these studies, digitization hardware, radiographic technique, normalization methods, window selection, and contrast medium distribution all contributed to differences in the statistics separating poor from good patient survival.

Original languageEnglish (US)
Pages (from-to)401-406
Number of pages6
JournalMedical Physics
Volume15
Issue number3
DOIs
StatePublished - 1988
Externally publishedYes

Fingerprint

Renal Cell Carcinoma
Cell Survival
Survival
Contrast Media
Lasers
Databases
Neoplasms

ASJC Scopus subject areas

  • Biophysics

Cite this

Medical pattern recognition pitfalls in a clinical setting : Renal cell carcinoma survival prediction. / Kimme-Smith, C.; Cochran, S. T.; Roe, Denise.

In: Medical Physics, Vol. 15, No. 3, 1988, p. 401-406.

Research output: Contribution to journalArticle

@article{ea500d441ff7429bab61f608fd08ba88,
title = "Medical pattern recognition pitfalls in a clinical setting: Renal cell carcinoma survival prediction",
abstract = "Three studies to predict renal cell carcinoma patient survival from tumor textures found in digitized early venous phase arteriograms were successful individually. However, when the three methods were compared, they were not consistent and no single method was clinically useful. The first study predicted 5 yr survival of 37 patients with 87{\%} accuracy. The second study added 29 patients to the data base; the poor survival of the 21 patients who died within 5 yr of diagnosis was predicted with 80.9{\%} accuracy. When 27 of these cases were redigitized with a laser scanner, average survival prediction accuracy was 78{\%}. In these studies, digitization hardware, radiographic technique, normalization methods, window selection, and contrast medium distribution all contributed to differences in the statistics separating poor from good patient survival.",
author = "C. Kimme-Smith and Cochran, {S. T.} and Denise Roe",
year = "1988",
doi = "10.1118/1.596265",
language = "English (US)",
volume = "15",
pages = "401--406",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "3",

}

TY - JOUR

T1 - Medical pattern recognition pitfalls in a clinical setting

T2 - Renal cell carcinoma survival prediction

AU - Kimme-Smith, C.

AU - Cochran, S. T.

AU - Roe, Denise

PY - 1988

Y1 - 1988

N2 - Three studies to predict renal cell carcinoma patient survival from tumor textures found in digitized early venous phase arteriograms were successful individually. However, when the three methods were compared, they were not consistent and no single method was clinically useful. The first study predicted 5 yr survival of 37 patients with 87% accuracy. The second study added 29 patients to the data base; the poor survival of the 21 patients who died within 5 yr of diagnosis was predicted with 80.9% accuracy. When 27 of these cases were redigitized with a laser scanner, average survival prediction accuracy was 78%. In these studies, digitization hardware, radiographic technique, normalization methods, window selection, and contrast medium distribution all contributed to differences in the statistics separating poor from good patient survival.

AB - Three studies to predict renal cell carcinoma patient survival from tumor textures found in digitized early venous phase arteriograms were successful individually. However, when the three methods were compared, they were not consistent and no single method was clinically useful. The first study predicted 5 yr survival of 37 patients with 87% accuracy. The second study added 29 patients to the data base; the poor survival of the 21 patients who died within 5 yr of diagnosis was predicted with 80.9% accuracy. When 27 of these cases were redigitized with a laser scanner, average survival prediction accuracy was 78%. In these studies, digitization hardware, radiographic technique, normalization methods, window selection, and contrast medium distribution all contributed to differences in the statistics separating poor from good patient survival.

UR - http://www.scopus.com/inward/record.url?scp=0023908771&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023908771&partnerID=8YFLogxK

U2 - 10.1118/1.596265

DO - 10.1118/1.596265

M3 - Article

C2 - 3405145

AN - SCOPUS:0023908771

VL - 15

SP - 401

EP - 406

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 3

ER -