Data-Driven Modelling and Control for Robot Needle Insertion in Deep Anterior Lamellar Keratoplasty

Edwards, William R.; Tang, Gao; Tian, Yuan; Draelos, Mark; Izatt, Joseph A.; Kuo, Anthony N.; Hauser, Kris

doi:10.1109/lra.2022.3140458

Cited by 12 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A more recent work proposes the usage of tactile sensors mounted on the fingers' tip to estimate the deformability of the wire to permit the insertion in a hollow cylinder [8] or in a control box component [9]. From the opposite perspective, other research aimed at the insertion of rigid object inside a flexible material [10], [11].…”

Section: Related Workmentioning

confidence: 99%

Cable Detection and Manipulation for DLO-in-Hole Assembly Tasks

Galassi

Caporali

Palli

2022

2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS)

View full text Add to dashboard Cite

This paper describes a cyber-physical system for the manipulation of Deformable Linear Objects (DLOs) addressing the DLO-in-hole insertion problem targeting an industrial scenario, the switchgear's components cabling task. In particular, the task considered is the insertion of DLOs in the switchgear components' holes. This task is very challenging since a precise knowledge of the DLO tip position and orientation is required for a successful operation. We tackled the DLO-in-hole problem from the computer vision perspective constraining our setup on employing just simple 2D images and by using the mobility of the robotic arm for achieving the full 3D knowledge of the DLOs. Then, the DLO tip is detected from two different image planes and the robot's trajectory corrected accordingly before insertion. To prove the effectiveness of the proposed solution, an example scenario is prepared and the method validated experimentally attempting the insertion of several DLOs in a sample switchgear component, obtaining an overall insertion success rate of 82.5%.

show abstract

Section: Related Workmentioning

confidence: 99%

Cable Detection and Manipulation for DLO-in-Hole Assembly Tasks

Galassi

Caporali

Palli

2022

2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS)

View full text Add to dashboard Cite

show abstract

“…Notably, [17] demonstrated OCT-guided vein cannulation using a silicone phantom. Others have demonstrated the feasibility of using iOCT for micro-suturing on a silicone phantom [18] and performing corneal kerotoplasty on ex-vivo human cornea [19]. While iOCT provides depth measurements that are sufficiently accurate for autonomous applications, a major drawback is its slow image acquisition and refresh rates due to the large amount of data processing involved.…”

Section: Related Workmentioning

confidence: 99%

Autonomously Navigating a Surgical Tool Inside the Eye by Learning from Demonstration

Kim

Urias

et al. 2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

A fundamental challenge in retinal surgery is safely navigating a surgical tool to a desired goal position on the retinal surface while avoiding damage to surrounding tissues, a procedure that typically requires tens-of-microns accuracy. In practice, the surgeon relies on depth-estimation skills to localize the tool-tip with respect to the retina in order to perform the tool-navigation task, which can be prone to human error. To alleviate such uncertainty, prior work has introduced ways to assist the surgeon by estimating the tooltip distance to the retina and providing haptic or auditory feedback. However, automating the tool-navigation task itself remains unsolved and largely unexplored. Such a capability, if reliably automated, could serve as a building block to streamline complex procedures and reduce the chance for tissue damage. Towards this end, we propose to automate the toolnavigation task by learning to mimic expert demonstrations of the task. Specifically, a deep network is trained to imitate expert trajectories toward various locations on the retina based on recorded visual servoing to a given goal specified by the user. The proposed autonomous navigation system is evaluated in simulation and in physical experiments using a silicone eye phantom. We show that the network can reliably navigate a needle surgical tool to various desired locations within 137 µm accuracy in physical experiments and 94 µm in simulation on average, and generalizes well to unseen situations such as in the presence of auxiliary surgical tools, variable eye backgrounds, and brightness conditions.

show abstract

“…Ophthalmic surgical robots at LoA 1 (robot assistance), with optical coherence tomography (OCT) guidance in subretinal injection [10] and deep anterior lamellar keratoplasty (DALK) [11] or with force sensing adaptation in retinal surgery [12], provide support for ophthalmologists but not control the operation. Surgical robots at LoA 2 (task autonomy) with elementary cognitive abilities, including depth tracking of needle tip [13], deformation evaluation in needle insertion [14] and trajectories learning in knot tying [15], execute a planned procedure but cannot update any parameter. LoA 3 (conditional autonomy) systems can extract parameters from the environment to plan a specific task.…”

mentioning

confidence: 99%

AIRS: Autonomous Intraoperative Robotic Suturing Based on Surgeon-Like Operation and Path Quantification in Keratoplasty

Feng,

Zhang,

Shi

et al. 2024

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

Automating surgical subtasks holds vast potential in robot-assisted surgeries, able to reduce surgeon fatigue and improve autonomy level. In this article, we propose an Autonomous Intraoperative Robotic Suturing (AIRS) system to imitate ophthalmologists to deliver a whole keratoplasty suturing including needle manipulation and knot tying. We introduce an autonomous strategy with a high-level task planner to generate quantified surgical paths and a low-level robotic controller to execute the paths. We first decompose the human keratoplasty into eight surgical phases and quantify the path primitives for each phase. Then two seven-DoF surgeon-like keratoplasty robots are built and the kinematics are deduced to execute surgical paths. To compute the changing motion parameters, a stereo vision algorithm is also developed to reconstruct 3-D features. Finally, to test the effectiveness of AIRS system, we perform autonomous robotic keratoplasty with single-interrupted sutures in phantom. With a desired threshold of 1 mm for position accuracy and 1 • for needle pose adaptation, AIRS is able to complete interrupted sutures at eight equally distributed suturing planes around the incision. Experimental results demonstrate the potential of AIRS system to accomplish complex tasks with full autonomy and improve surgical consistency.

show abstract

Data-Driven Modelling and Control for Robot Needle Insertion in Deep Anterior Lamellar Keratoplasty

Cited by 12 publications

References 42 publications

Cable Detection and Manipulation for DLO-in-Hole Assembly Tasks

Cable Detection and Manipulation for DLO-in-Hole Assembly Tasks

Autonomously Navigating a Surgical Tool Inside the Eye by Learning from Demonstration

AIRS: Autonomous Intraoperative Robotic Suturing Based on Surgeon-Like Operation and Path Quantification in Keratoplasty

Contact Info

Product

Resources

About