Artificial neural networks for non-invasive chromosomal abnormality screening of fetuses

Abstract

Ahstract-A large number of different neural network struc tures have been constructed, trained and tested to a large data base of pregnant women characteristics, aiming at generating a classifier-predictor for the presence of chromosomal abnor malities in fetuses, namely the Trisomy 21 (Down syndrome), Trisomy 18 (Edwards syndrome), Trisomy 13 (Patau syn drome) and the Turner syndrome. The database was composed of 31611 cases of pregnant women. 31135 women did not show any chromosomal abnor malities, while the remaining 476 were confirmed as having a chromosomal anomaly ofT21, T18, T13, or Turner Syndrome. From the total of 31611 cases, 8191 were kept as a totally unknown database that was only used for the verification of the predictability of the network. In this set, 7 were of the Turner syndrome, 14 of the Patau syndrome, 42 of the Edwards syn drome and 71 of the Down syndrome.For each subject, 10 parameters were considered to be the most influential at characterizing the risk of occurrence of these types of chromosomal anomalies.The best results were obtained when using a multi-layer neu ral structure having an input, an output and three hidden lay ers. For the case of the totally unknown verification set of the 8191 cases, 98.1 % were correctly identified. The percentage of abnormal cases correctly predicted was 85.1 %. The unknown T21 cases were predicted by 78.9%, the T18 by 76.2%, the T13 by 0.0% and the Turner syndrome by 42.9%.

First Trimester Noninvasive Prenatal Diagnosis: A Computational Intelligence Approach

“…The computational artificial neural networks (ANNs), which are a specific paradigm of computational intelligence, had been used as effective classifiers and predictors for the last 25 years. Indeed, they had been extensively applied in medical and biological research and applications [17]- [20].…”

Down syndrome identification and classification using facial features with neural network

“…Loos et al proposed a computer-based graph matching algorithm for DS analysis. Artificial Neural Network based automated methods are proposed by many authors [10,11] to recognize and detect the fetal nasal bone.…”

Artificial Neural Networks to Investigate the Importance and the Sensitivity to Various Parameters Used for the Prediction of Chromosomal Abnormalities

“…It is known that for an effective non-invasive first-trimester screening for fetal chromosomal abnormalities, various maternal and feto-placental parameters may be explored in an appropriate diagnostic tool ( [1], [2], [7]). The most important parameters for diagnosing chromosomal abnormalities are the maternal age (MA), the previous occurrence of aneuploidies, the fetal nuchal translucency (NT) thickness, the crown rump length (CRL), the free β-human chorionic gonadotrophin (b-hCG), the pregnancy associated plasma protein-A (PAPP-A), the nasal bone (NB), the tricuspid flow (TF) and the ductus venosus flow (DV) [2].…”

“…In a previous study [7], an accurate system for the non-invasive identification of trisomy 21 during the first three months of the pregnancy was suggested. That system was tested with 129 unknown cases of T21 and 16,898 unknown cases of normal fetuses.…”