Predicting Blastocyst Formation in IVF: Integrating DINOv2 and Attention-Based LSTM on Time-Lapse Embryo Images

2026-04-14

AI Summary

This study addresses the challenge of selecting high-quality embryos in in vitro fertilization (IVF) when limited to sparse daily images or incomplete time-lapse videos, which hinders reliable blastocyst formation prediction. To overcome this, the authors propose a novel temporal modeling approach that integrates the DINOv2 vision model with a multi-head attention-enhanced LSTM architecture. This is the first work to combine DINOv2 and attention-augmented LSTM for embryo development prediction, enabling efficient extraction of spatiotemporal features from incomplete time-lapse image sequences. Evaluated on a dataset of 704 embryos, the method achieves 96.4% accuracy—significantly outperforming existing approaches—and demonstrates robust performance even under substantial frame missingness, thereby enhancing clinical applicability and supporting embryologists’ decision-making efficiency.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The selection of the optimal embryo for transfer is a critical yet challenging step in in vitro fertilization (IVF), primarily due to its reliance on the manual inspection of extensive time-lapse imaging data. A key obstacle in this process is predicting blastocyst formation from the limited number of daily images available. Many clinics also lack complete time-lapse systems, so full videos are often unavailable. In this study, we aimed to predict which embryos will develop into blastocysts using limited daily images from time-lapse recordings. We propose a novel hybrid model that combines DINOv2, a transformer-based vision model, with an enhanced long short-term memory (LSTM) network featuring a multi-head attention layer. DINOv2 extracts meaningful features from embryo images, and the LSTM model then uses these features to analyze embryo development over time and generate final predictions. We tested our model on a real dataset of 704 embryo videos. The model achieved 96.4% accuracy, surpassing existing methods. It also performs well with missing frames, making it valuable for many IVF laboratories with limited imaging systems. Our approach can assist embryologists in selecting better embryos more efficiently and with greater confidence.

Problem

• blastocyst formation prediction
• time-lapse embryo imaging
• embryo selection
• in vitro fertilization (IVF)
• limited imaging data

Innovation

• DINOv2
• Attention-based LSTM
• Blastocyst prediction
• Time-lapse embryo imaging
• Embryo selection