Automated trisomy 21 assessment based on maternal serum markers using trivariate lognormal distribution

Authors:
Lai Khin Wee;Lim Miin;Eko Supriyanto
Affiliations:
Department of Clinical Science and Engineering, Faculty of Health Science and Biomedical Engineering, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia;Department of Clinical Science and Engineering, Faculty of Health Science and Biomedical Engineering, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia;Department of Clinical Science and Engineering, Faculty of Health Science and Biomedical Engineering, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia
Venue:
WSEAS TRANSACTIONS on SYSTEMS
Year:
2010

Citing 4
Cited 0

Inference in multi-state survival data

MATH'08 Proceedings of the 13th WSEAS international conference on Applied mathematics
A penalized fuzzy clustering algorithm

ACS'06 Proceedings of the 6th WSEAS international conference on Applied computer science
Wavelet-based statistical analysis in functional neuroimaging

WAMUS'06 Proceedings of the 6th WSEAS international conference on Wavelet analysis & multirate systems
Feature extraction in digital mammography using LabVIEW

CONTROL'05 Proceedings of the 2005 WSEAS international conference on Dynamical systems and control

Quantified Score

Hi-index	0.00

Visualization

Abstract

Trisomy 21 is the most frequent types of chromosomal abnormalities. Generally, current methods for trisomy risk assessment are divided into two techniques, which are invasive and non-invasive methods. Invasive methods are including amniocentesis, chorionic villus sampling (CVS), or percutaneous umbilical cord blood sampling (PUBS), but its drawbacks are expensive, time consuming and having risk of miscarriage, where else non-invasive methods are counting on ultrasound marker and maternal serum markers screening. Nevertheless, single evaluations on ultrasound markers itself are always not enough for risk assessment in terms of its accuracy, reliability and consistency. So, we proposed a new mathematical algorithm which combines three maternal serum markers using trivariate lognormal distribution to calculate automatically the probability or likelihood that a woman has an affected pregnancy or not. The developed algorithm was implemented into graphical user interface to act as computer aided e-diagnostic system. We have compared the results with published finding and found it is almost equally accurate.