Structured-Light-Based Robotic System with Pre-Fixation for Automated Intravitreal Injection.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 41805526.
- Also identified by DOI 10.1109/TBME.2026.3672601.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Intravitreal injection (IVI), a critical treatment for ophthalmic diseases requiring precise scleral puncture, is traditionally performed manually, demanding high skill, limiting efficiency, and lacking standardization. This study aims to develop and evaluate an automated IVI (A-IVI) robotic system to improve accuracy, efficiency, and safety. The proposed system comprises a custom-designed compact injection robot, a structured-light camera, and a general robotic arm. Based on the optimal injection pose determined from point clouds acquired by the vision sensor, the injection robot performs IVI. Before needle insertion, the eyeball is pre-fixed using a compliance-controlled fixture to enhance safety and comfort. A motion-planning strategy coordinates the degrees of freedom (DOFs) of the remote-center-of-motion (RCM) mechanism and robotic arm, enabling safe and flexible needle puncture. Experiments were conducted on eye phantoms and ex-vivo porcine eyes. Mean execution times were 38.3 s and 40.8 s, puncture accuracy was 3.79±0.31 mm and 3.68±0.83 mm, and compliant interaction forces ranged from 0.25-1.15 N and 0.18-0.98 N, respectively, all within the clinically acceptable range. The proposed robotic system achieves clinically acceptable accuracy and safe interaction forces while reducing variability and standardizing the IVI procedure. This is the first application of structured-light-based measurement to IVI, enabling automation with precision and safety. The system demonstrates potential to improve efficiency, reduce surgeon workload, and establish a standardized approach to IVI in clinical practice.