2020/05/21 by Prudhvi Thirumalaraju, Manoj Kumar Kanakasabapathy, Thirumalaraju, Prudhvi +15 · 1 citation
Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Reproductive Biology and Fertility #Sperm and Testicular Function #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.10912
openalex publication_date 2020/05/21 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
A critical factor that influences the success of an in-vitro fertilization\n(IVF) procedure is the quality of the transferred embryo. Embryo morphology\nassessments, conventionally performed through manual microscopic analysis\nsuffer from disparities in practice, selection criteria, and subjectivity due\nto the experience of the embryologist. Convolutional neural networks (CNNs) are\npowerful, promising algorithms with significant potential for accurate\nclassifications across many object categories. Network architectures and\nhyper-parameters affect the efficiency of CNNs for any given task. Here, we\nevaluate multi-layered CNNs developed from scratch and popular deep-learning\narchitectures such as Inception v3, ResNET, Inception-ResNET-v2, and Xception\nin differentiating between embryos based on their morphological quality at 113\nhours post insemination (hpi). Xception performed the best in differentiating\nbetween the embryos based on their morphological quality.\n