Semantic Video Segmentation for Intracytoplasmic Sperm Injection Procedures
This work addresses the problem of automating analysis of ICSI procedures for embryologists, potentially improving efficiency and consistency in fertility treatments.
This paper introduces the first deep learning model for analyzing intracytoplasmic sperm injection (ICSI) procedures. The model achieves a mean IoU of 0.962 for segmenting key objects and a mean pixel error of 3.793 for localizing the needle tip, running at 14 FPS.
We present the first deep learning model for the analysis of intracytoplasmic sperm injection (ICSI) procedures. Using a dataset of ICSI procedure videos, we train a deep neural network to segment key objects in the videos achieving a mean IoU of 0.962, and to localize the needle tip achieving a mean pixel error of 3.793 pixels at 14 FPS on a single GPU. We further analyze the variation between the dataset's human annotators and find the model's performance to be comparable to human experts.