Related Experiment Video
Updated: Jan 16, 2026
10:10
Ultrahigh Resolution Mouse Optical Coherence Tomography to Aid Intraocular Injection in Retinal Gene Therapy Research
Published on: November 2, 2018
9.8K
Deep Learning-Enhanced Robotic Subretinal Injection with Real-Time Retinal Motion Compensation
Tianle Wu1, Mojtaba Esfandiari1, Peiyao Zhang1
1Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, MD, USA.
IEEE International Conference on Automation Science and Engineering (CASE) : [Proceedings]. IEEE Conference on Automation Science and Engineering
|September 26, 2025
Summary
This study introduces an AI-powered robotic system for subretinal injections, improving precision in treating retinal diseases. It synchronizes needle and retinal motion using deep learning, enhancing surgical safety and accuracy.
Area of Science:
- Ophthalmology
- Robotics
- Artificial Intelligence
Background:
- Subretinal injection is crucial for treating retinal diseases like inherited retinal diseases (IRD) and age-related macular degeneration (AMD).
- Physiological retinal motion poses challenges for precise needle placement, risking retinal pigment epithelium (RPE) damage.
Purpose of the Study:
- To develop a fully autonomous robotic system for subretinal injection.
- To enhance surgical precision and safety by synchronizing needle and retinal motion.
Main Methods:
- Integration of intraoperative optical coherence tomography (iOCT) imaging and deep learning for motion prediction.
- Utilizing a Long Short-Term Memory (LSTM) neural network to predict internal limiting membrane (ILM) motion.
- Implementation of a real-time registration framework and dynamic proportional speed control for needle insertion.
Main Results:
- The LSTM model outperformed a Fast Fourier Transform (FFT) baseline in predicting ILM motion.
- Experimental validation in simulations and ex vivo porcine eyes demonstrated precise motion synchronization.
- Achieved a mean tracking error below 16.4 μm in pre-insertion phases.
Conclusions:
- The AI-driven robotic system shows significant potential for improving the safety and accuracy of retinal microsurgery.
- Autonomous robotic assistance can overcome challenges posed by physiological retinal motion.
