|
Title
|
|
|
|
Selecting the embryo with the highest implantation potential using a data mining based prediction model
| |
|
Author
|
|
|
|
| |
|
Abstract
|
|
|
|
| Background: Embryo selection has been based on developmental and morphological characteristics. However, the presence of an important intra- and inter-observer variability of standard scoring system (SSS) has been reported. A computer-assisted scoring system (CASS) has the potential to overcome most of these disadvantages associated with the SSS. The aims of this study were to construct a prediction model, with data mining approaches, and compare the predictive performance of models in SSS and CASS and to evaluate whether using the prediction model would impact the selection of the embryo for transfer. Methods: A total of 871 single transferred embryos between 2008 and 2013 were included and evaluated with two scoring systems: SSS and CASS. Prediction models were developed using multivariable logistic regression (LR) and multivariate adaptive regression splines (MARS). The prediction models were externally validated with a test set of 109 single transfers between January and June 2014. Area under the curve (AUC) in training data and validation data was compared to determine the utility of the models. Results: In SSS models, the AUC declined significantly from training data to validation data (p < 0.05). No significant difference was detected in CASS derived models. Two final prediction models derived from CASS were obtained using LR and MARS, which showed moderate discriminative capacity (c-statistic 0.64 and 0.69 respectively) on validation data. Conclusions: The study showed that the introduction of CASS improved the generalizability of the prediction models, and the combination of computer-assisted scoring system with data mining based predictive modeling is a promising approach to improve the selection of embryo with the highest implantation potential. |
| |
|
Language
|
|
|
|
English
| |
|
Source (journal)
|
|
|
|
Reproductive biology and endocrinology. - London
| |
|
Publication
|
|
|
|
London
:
2016
| |
|
ISSN
|
|
|
|
1477-7827
| |
|
DOI
|
|
|
|
10.1186/S12958-016-0145-1
| |
|
Volume/pages
|
|
|
|
14
(2016)
, 12 p.
| |
|
Article Reference
|
|
|
|
10
| |
|
ISI
|
|
|
|
000371590400001
| |
|
Medium
|
|
|
|
E-only publicatie
| |
|
Full text (Publisher's DOI)
|
|
|
|
| |
|
Full text (open access)
|
|
|
|
| |
|