@inproceedings{07f31588d6f34ff1ad8ff858c3da62bd,
title = "Diagnosis of iron-deficiency anemia in hemodialyzed patients through support vector machines technique",
abstract = "Support Vector Machines (SVMs) technique is a recent method for empirical data modelling applied to pattern recognition problems. The aim of the present study is to test SVMs performance when applied to a specific medical classification problem – diagnosis of iron-deficiency anemia in uremic patients - and to compare the results with those obtained by traditional techniques such as logistic regression and discriminant analysis. Models have been compared both in learning and validation phases. All methods performed well (accuracy > 80%). Sensibility of SVMs is always higher than the ones of the other models; specificity and accuracy are lower in one repetition over three. Within the limits of the present study, we can say that the SVMs can constitute an innovative method to approach clinical classification problem on which to further invest.",
author = "Paola Baiardi and Valter Piazza and Mazzoleni, {Maria C.}",
year = "2001",
language = "English",
isbn = "3540422943",
volume = "2101",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "144--147",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
note = "8th Conference on Artificial Intelligence in Medicine in Europe, AIME 2001 ; Conference date: 01-07-2001 Through 04-07-2001",
}