Expert-Annotated Embryo Image Dataset with Natural Language Descriptions for Evidence-Based Patient Communication in IVF

arXiv cs.CV / 4/21/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes an expert-annotated IVF embryo image dataset paired with natural-language morphological descriptions to support evidence-based embryo selection and clearer patient communication.
It argues that prior AI approaches have limited impact because they must be adapted to specific clinical data, often depend on time-lapse incubators, and provide insufficient interpretability of AI reasoning.
The dataset includes descriptions covering relevant biological aspects such as embryonic cell cycle, developmental stage, and morphological features.
The authors suggest using the dataset to fine-tune vision-language foundation models, generate predicted embryo descriptions, and then extract supporting scientific evidence from literature for transparent decision justification.
They position the dataset as a foundation for research into language-based, interpretable, and transparent automated embryo assessment that could improve decision-making and outcomes over time.

Abstract

Embryo selection is one of multiple crucial steps in in-vitro fertilization, commonly based on morphological assessment by clinical embryologists. Although artificial intelligence methods have demonstrated their potential to support embryo selection by automated embryo ranking or grading methods, the overall impact of AI-based solutions is still limited. This is mainly due to the required adaptation of automated solutions to custom clinical data, reliance on time lapse incubators and a lack of interpretability to understand AI reasoning. The modern, informed patient is questioning expert decisions, particularly if the treatment is not successful. Thus, evidence-based decision justification in tasks like embryo selection would support transparent decision making and respectful patient communication. To support this aim, we hereby present an expert-annotated dataset consisting of embryo images and corresponding morphological description using natural language. The description contains relevant information on embryonic cell cycle, developmental stage and morphological features. This dataset enables the finetuning of modern foundational vision-language models to learn and improve over time with high accuracy. Predicted embryo descriptions can then be leveraged to automatically extract scientific evidence from literature, facilitating well-informed, evidence-based decision-making and transparent communication with patients. Our proposed dataset supports research in language-based, interpretable, and transparent automated embryo assessment and has the potential to enhance the decision-making process and improve patient outcomes significantly over time.