Deep Learning-Enhanced Robotic Subretinal Injection with Real-Time Retinal Motion Compensation

2025-04-04

AI Summary

Subretinal injection for diseases such as age-related macular degeneration (AMD) suffers from targeting inaccuracies caused by physiological retinal motion induced by respiration and cardiac pulsation, thereby increasing the risk of retinal pigment epithelium (RPE) damage. To address this, we propose a fully automated robotic system that integrates intraoperative optical coherence tomography (iOCT) imaging with an LSTM-based real-time internal limiting membrane (ILM) motion prediction model to dynamically compensate for respiration- and heartbeat-induced displacements. Our approach introduces two key innovations: (1) real-time iOCT–robot coordinate system registration, and (2) dynamic scaling velocity control—overcoming the accuracy limitations of conventional open-loop and FFT-based methods. In ex vivo porcine eye experiments, the system achieved a mean tracking error of <16.4 μm and successfully performed high-precision subretinal injections without RPE injury, demonstrating clinical feasibility.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Subretinal injection is a critical procedure for delivering therapeutic agents to treat retinal diseases such as age-related macular degeneration (AMD). However, retinal motion caused by physiological factors such as respiration and heartbeat significantly impacts precise needle positioning, increasing the risk of retinal pigment epithelium (RPE) damage. This paper presents a fully autonomous robotic subretinal injection system that integrates intraoperative optical coherence tomography (iOCT) imaging and deep learning-based motion prediction to synchronize needle motion with retinal displacement. A Long Short-Term Memory (LSTM) neural network is used to predict internal limiting membrane (ILM) motion, outperforming a Fast Fourier Transform (FFT)-based baseline model. Additionally, a real-time registration framework aligns the needle tip position with the robot's coordinate frame. Then, a dynamic proportional speed control strategy ensures smooth and adaptive needle insertion. Experimental validation in both simulation and ex vivo open-sky porcine eyes demonstrates precise motion synchronization and successful subretinal injections. The experiment achieves a mean tracking error below 16.4 {mu}m in pre-insertion phases. These results show the potential of AI-driven robotic assistance to improve the safety and accuracy of retinal microsurgery.

Problem

• Autonomous robotic system for precise subretinal injection
• Compensates retinal motion caused by heartbeat and respiration
• Uses deep learning and iOCT for real-time motion prediction

Innovation

• Deep learning predicts retinal motion for synchronization
• Real-time registration aligns needle with robot coordinates
• Dynamic speed control ensures smooth needle insertion