Premium
Quantitative thermography as a predictor of breast cancer
Author(s) -
Barash I. M.,
Pasternack B. S.,
Venet L.,
Wolff W. I.
Publication year - 1973
Publication title -
cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.052
H-Index - 304
eISSN - 1097-0142
pISSN - 0008-543X
DOI - 10.1002/1097-0142(197304)31:4<769::aid-cncr2820310402>3.0.co;2-7
Subject(s) - medicine , weighting , thermography , breast cancer , linear discriminant analysis , discriminant function analysis , cancer , malignancy , mammography , statistics , radiology , mathematics , physics , infrared , optics
The lack of specific, consistent diagnostic criteria has lessened the value of thermography in the diagnosis of mammary cancer. In an effort to improve the diagnostic accuracy of this technique, a numerical procedure was devised based on the evaluation of four features of the breast thermogram. Each was shown to be a statistically valid indicator of malignancy. In combination, as a diagnostic score, they were even more reliable. The development of a color technique provided a simple means for quantitating three of these features. Normal patients and patients with proven breast tumors were thermographed and scored. Optimal weighting, by linear discriminant analysis, did not improve the results significantly. Three sets of weights for the four diagnostic criteria were used in the multiple logistic function in order to generate probability tables which related all possible diagnostic scores to the chance of having cancer. The observed results, in this small series, compared favorably with those predicted by the model.