We describe the steps of our machine learning pipeline to phenotype oocyte based on images/movies acquired noninvasively in transmitted light. The pipeline allows to extract hundreds of morphological/dynamical features describing individual oocytes by segmenting the oocyte contours and measuring the corresponding features. Oocytes are then characterized by running feature-based machine learning algorithm on the features.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Machine Learning Methods to Characterize Mouse and Human Oocytes

  • Gaëlle Letort

摘要

We describe the steps of our machine learning pipeline to phenotype oocyte based on images/movies acquired noninvasively in transmitted light. The pipeline allows to extract hundreds of morphological/dynamical features describing individual oocytes by segmenting the oocyte contours and measuring the corresponding features. Oocytes are then characterized by running feature-based machine learning algorithm on the features.