Paraskevi Vogiatzi1, Abraham Pouliakis2, Charalampos Siristatidis3. 1. Assisted Reproduction Unit, Third Department of Obstetrics and Gynecology, Medical School, "Attikon" University Hospital, National and Kapodistrian University of Athens, 1 Rimini Street, Chaidari, 12642, Athens, Greece. 2. Second Department of Pathology, Medical School, "Attikon" University Hospital, National and Kapodistrian University of Athens, 1 Rimini Street, Chaidari, 12642, Athens, Greece. 3. Assisted Reproduction Unit, Third Department of Obstetrics and Gynecology, Medical School, "Attikon" University Hospital, National and Kapodistrian University of Athens, 1 Rimini Street, Chaidari, 12642, Athens, Greece. harrysiri@yahoo.gr.
Abstract
PURPOSE: To construct and validate an efficient artificial neural network (ANN) based on parameters with statistical correlation to live birth, to be used as a comprehensive tool for the prediction of the clinical outcome for patients undergoing ART. METHODS: Data from 257 infertile couples that underwent a total of 426 IVF/ICSI cycles from 2010 to 2017 was collected on an ensemble of 118 parameters for each cycle. Statistical correlation of the parameters with the outcome of live birth was performed, using either t test or χ2 test, and the parameters that demonstrated statistical significance were used to construct the ANN. Cross-validation was performed by random separation of data and repeating the training-testing procedure by 10 times. RESULTS: 12 statistically significant parameters out of the initial ensemble were used for the ANN construction, which exhibited a cumulative sensitivity and specificity of 76.7% and 73.4%, respectively. During cross-validation, the system exhibited the following: sensitivity 69.2% ± 2.36%, specificity 69.19% ± 2.8% (OR 5.21 ± 1.27), PPV 36.96 ± 3.44, NPV 89.61 ± 1.09, and OA 69.19% ± 2.69%. A rather small standard deviation in the performance indices between the training and test sets throughout the validation process indicated a stable performance of the constructed ANN. CONCLUSIONS: The constructed ANN is based on statistically significant variables with the outcome of live birth and represents a stable and efficient system with increased performance indices. Validation of the system allowed an insight of its clinical value as a supportive tool in medical decisions, and overall provides a reliable approach in the routine practice of IVF units in a user-friendly environment.
PURPOSE: To construct and validate an efficient artificial neural network (ANN) based on parameters with statistical correlation to live birth, to be used as a comprehensive tool for the prediction of the clinical outcome for patients undergoing ART. METHODS: Data from 257 infertile couples that underwent a total of 426 IVF/ICSI cycles from 2010 to 2017 was collected on an ensemble of 118 parameters for each cycle. Statistical correlation of the parameters with the outcome of live birth was performed, using either t test or χ2 test, and the parameters that demonstrated statistical significance were used to construct the ANN. Cross-validation was performed by random separation of data and repeating the training-testing procedure by 10 times. RESULTS: 12 statistically significant parameters out of the initial ensemble were used for the ANN construction, which exhibited a cumulative sensitivity and specificity of 76.7% and 73.4%, respectively. During cross-validation, the system exhibited the following: sensitivity 69.2% ± 2.36%, specificity 69.19% ± 2.8% (OR 5.21 ± 1.27), PPV 36.96 ± 3.44, NPV 89.61 ± 1.09, and OA 69.19% ± 2.69%. A rather small standard deviation in the performance indices between the training and test sets throughout the validation process indicated a stable performance of the constructed ANN. CONCLUSIONS: The constructed ANN is based on statistically significant variables with the outcome of live birth and represents a stable and efficient system with increased performance indices. Validation of the system allowed an insight of its clinical value as a supportive tool in medical decisions, and overall provides a reliable approach in the routine practice of IVF units in a user-friendly environment.
Entities:
Keywords:
Artificial intelligence; Artificial neural network; Assisted reproduction; Live birth; Personalized treatment; Prediction model
Authors: Prajna Banerjee; Bokyung Choi; Lora K Shahine; Sunny H Jun; Kathleen O'Leary; Ruth B Lathi; Lynn M Westphal; Wing H Wong; Mylene W M Yao Journal: Proc Natl Acad Sci U S A Date: 2010-07-19 Impact factor: 11.205
Authors: Valerie L Baker; Barbara Luke; Morton B Brown; Ruben Alvero; John L Frattarelli; Rebecca Usadi; David A Grainger; Alicia Y Armstrong Journal: Fertil Steril Date: 2009-09-09 Impact factor: 7.329
Authors: J de Mouzon; V Goossens; S Bhattacharya; J A Castilla; A P Ferraretti; V Korsak; M Kupka; K G Nygren; A Nyboe Andersen Journal: Hum Reprod Date: 2010-06-22 Impact factor: 6.918