Scholarlink

Unsupervised algorithms are valuable in medicine for discovering hidden patterns without relying on predefined labels. In this work, we conduct an experimental study to test different models on a novel dataset for classifying oocytes as viable or non-viable. The effectiveness of the models is evaluated by measuring the Area Under the Curve (AUC), which is comparable to supervised works. In addition to this, heatmaps have been extracted to provide explainability to the models, so a qualitative analysis has been carried out.

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

Unsupervised Prediction of Blastocyst Development from Oocyte Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Natalia Pérez García-de-la-Puente,
  • Elena Paya,
  • Lucía Murria,
  • Miguel López-Pérez,
  • Marcos Meseguer,
  • Valery Naranjo

摘要

Unsupervised algorithms are valuable in medicine for discovering hidden patterns without relying on predefined labels. In this work, we conduct an experimental study to test different models on a novel dataset for classifying oocytes as viable or non-viable. The effectiveness of the models is evaluated by measuring the Area Under the Curve (AUC), which is comparable to supervised works. In addition to this, heatmaps have been extracted to provide explainability to the models, so a qualitative analysis has been carried out.