Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines

Sung-Huai Hsieh; Yin-Hsiu Chien; Chia-Ping Shen; Wei-Hsin Chen; Po Hao Chen; Sheau-Ling Hsieh; Po-Hsun Cheng; Feipei Lai

doi:10.1109/BIBE.2009.72

Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines

Sung-Huai Hsieh, Yin-Hsiu Chien, Chia-Ping Shen, Wei-Hsin Chen, Po Hao Chen, Sheau-Ling Hsieh, Po-Hsun Cheng, Feipei Lai

Center for Information Technology Services

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

The clinical symptoms of metabolic disorders during neonatal period are often not apparent, if not treated early irreversible damages such as mental retardation may occur, even death. Therefore, practicing newborn screening is very important to prevent neonatal from these damages. In this paper, the newborn screening system used support vector machines (SVM) classification technique is proposed in place of cut-off value decision to evaluate the metabolic substances concentration raw data obtained from tandem mass spectrometry (MS/MS) and determine whether the newborn has some kinds of metabolic disorder diseases. On the basis of the proposed features, new analytic combinations are identified with superior discriminatory performance compared with the best published combinations. Classifiers built with the feature selection to find C3/C2, C3 and C16 of three key point features achieved diagnostic sensitivities, specificities and accuracy approaching 100%.

Original language	English
Title of host publication	9th IEEE International Conference on BioInformatics and BioEngineering
Publisher	IEEE
DOIs	https://doi.org/10.1109/BIBE.2009.72
State	Published - 2009

Keywords

TANDEM MASS-SPECTROMETRY

Access to Document

10.1109/BIBE.2009.72

Fingerprint Dive into the research topics of 'Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines'. Together they form a unique fingerprint.

Cite this

Hsieh, S-H., Chien, Y-H., Shen, C-P., Chen, W-H., Chen, P. H., Hsieh, S-L., Cheng, P-H., & Lai, F. (2009). Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines. In 9th IEEE International Conference on BioInformatics and BioEngineering IEEE. https://doi.org/10.1109/BIBE.2009.72

@inproceedings{0a0df30dcaf944e79f62dc0aa4995ccb,

title = "Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines",

abstract = "The clinical symptoms of metabolic disorders during neonatal period are often not apparent, if not treated early irreversible damages such as mental retardation may occur, even death. Therefore, practicing newborn screening is very important to prevent neonatal from these damages. In this paper, the newborn screening system used support vector machines (SVM) classification technique is proposed in place of cut-off value decision to evaluate the metabolic substances concentration raw data obtained from tandem mass spectrometry (MS/MS) and determine whether the newborn has some kinds of metabolic disorder diseases. On the basis of the proposed features, new analytic combinations are identified with superior discriminatory performance compared with the best published combinations. Classifiers built with the feature selection to find C3/C2, C3 and C16 of three key point features achieved diagnostic sensitivities, specificities and accuracy approaching 100%.",

keywords = "TANDEM MASS-SPECTROMETRY",

author = "Sung-Huai Hsieh and Yin-Hsiu Chien and Chia-Ping Shen and Wei-Hsin Chen and Chen, {Po Hao} and Sheau-Ling Hsieh and Po-Hsun Cheng and Feipei Lai",

year = "2009",

doi = "10.1109/BIBE.2009.72",

language = "English",

booktitle = "9th IEEE International Conference on BioInformatics and BioEngineering",

publisher = "IEEE",

}

Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines. / Hsieh, Sung-Huai; Chien, Yin-Hsiu; Shen, Chia-Ping; Chen, Wei-Hsin; Chen, Po Hao; Hsieh, Sheau-Ling; Cheng, Po-Hsun; Lai, Feipei.

9th IEEE International Conference on BioInformatics and BioEngineering. IEEE, 2009.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Newborn Screening System Based on Adaptive Feature Selection and Support Vector Machines

AU - Hsieh, Sung-Huai

AU - Chien, Yin-Hsiu

AU - Shen, Chia-Ping

AU - Chen, Wei-Hsin

AU - Chen, Po Hao

AU - Hsieh, Sheau-Ling

AU - Cheng, Po-Hsun

AU - Lai, Feipei

PY - 2009

Y1 - 2009

N2 - The clinical symptoms of metabolic disorders during neonatal period are often not apparent, if not treated early irreversible damages such as mental retardation may occur, even death. Therefore, practicing newborn screening is very important to prevent neonatal from these damages. In this paper, the newborn screening system used support vector machines (SVM) classification technique is proposed in place of cut-off value decision to evaluate the metabolic substances concentration raw data obtained from tandem mass spectrometry (MS/MS) and determine whether the newborn has some kinds of metabolic disorder diseases. On the basis of the proposed features, new analytic combinations are identified with superior discriminatory performance compared with the best published combinations. Classifiers built with the feature selection to find C3/C2, C3 and C16 of three key point features achieved diagnostic sensitivities, specificities and accuracy approaching 100%.

AB - The clinical symptoms of metabolic disorders during neonatal period are often not apparent, if not treated early irreversible damages such as mental retardation may occur, even death. Therefore, practicing newborn screening is very important to prevent neonatal from these damages. In this paper, the newborn screening system used support vector machines (SVM) classification technique is proposed in place of cut-off value decision to evaluate the metabolic substances concentration raw data obtained from tandem mass spectrometry (MS/MS) and determine whether the newborn has some kinds of metabolic disorder diseases. On the basis of the proposed features, new analytic combinations are identified with superior discriminatory performance compared with the best published combinations. Classifiers built with the feature selection to find C3/C2, C3 and C16 of three key point features achieved diagnostic sensitivities, specificities and accuracy approaching 100%.

KW - TANDEM MASS-SPECTROMETRY

U2 - 10.1109/BIBE.2009.72

DO - 10.1109/BIBE.2009.72

M3 - Conference contribution

BT - 9th IEEE International Conference on BioInformatics and BioEngineering

PB - IEEE

ER -