Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training

Paul Sullivan; Walid Shaker; Mustafa Suphi Erden

doi:10.1109/icarm65671.2025.11293476

Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training

Paul Sullivan
, Walid Shaker
, Mustafa Suphi Erden

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

24 Downloads (Pure)

Abstract

Laparoscopic approaches to surgery have become widely adopted, due to several advantages over open surgery. However, as an unavoidable consequence of the nature of the laparoscopic setup, skills are more difficult to acquire and sufficient quality feedback on simulator training is costly to procure. We propose a real-time algorithm to track instrument trajectories in 3D within traditional physical laparoscopic simulators. This enables feedback and performance assessment via machine learning techniques. The algorithm has been characterised using a robotically controlled laparoscopic setup as a ground truth.

Original language	English
Title of host publication	2025 IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)
Publisher	IEEE
ISBN (Electronic)	9798331503079
DOIs	https://doi.org/10.1109/icarm65671.2025.11293476
Publication status	Published - 22 Dec 2025
Event	10th IEEE International Conference on Advanced Robotics and Mechatronics 2025 - Portsmouth, United Kingdom Duration: 1 Aug 2025 → 3 Aug 2025 http://www.ieee-arm.org/

Conference

Conference	10th IEEE International Conference on Advanced Robotics and Mechatronics 2025
Abbreviated title	ARM 2025
Country/Territory	United Kingdom
City	Portsmouth
Period	1/08/25 → 3/08/25
Internet address	http://www.ieee-arm.org/

Access to Document

10.1109/icarm65671.2025.11293476

Download pdfAccepted author manuscript, 2.13 MBLicence: CC BY

Cite this

@inproceedings{700d8a083d2d4ca4b81d14e64c314887,

title = "Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training",

abstract = "Laparoscopic approaches to surgery have become widely adopted, due to several advantages over open surgery. However, as an unavoidable consequence of the nature of the laparoscopic setup, skills are more difficult to acquire and sufficient quality feedback on simulator training is costly to procure. We propose a real-time algorithm to track instrument trajectories in 3D within traditional physical laparoscopic simulators. This enables feedback and performance assessment via machine learning techniques. The algorithm has been characterised using a robotically controlled laparoscopic setup as a ground truth.",

author = "Paul Sullivan and Walid Shaker and Erden, \{Mustafa Suphi\}",

year = "2025",

month = dec,

day = "22",

doi = "10.1109/icarm65671.2025.11293476",

language = "English",

booktitle = "2025 IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)",

publisher = "IEEE",

address = "United States",

note = "10th IEEE International Conference on Advanced Robotics and Mechatronics 2025, ARM 2025 ; Conference date: 01-08-2025 Through 03-08-2025",

url = "http://www.ieee-arm.org/",

}

Sullivan, P, Shaker, W & Erden, MS 2025, Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training. in 2025 IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)., 11293476, IEEE, 10th IEEE International Conference on Advanced Robotics and Mechatronics 2025, Portsmouth, United Kingdom, 1/08/25. https://doi.org/10.1109/icarm65671.2025.11293476

TY - GEN

T1 - Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training

AU - Sullivan, Paul

AU - Shaker, Walid

AU - Erden, Mustafa Suphi

PY - 2025/12/22

Y1 - 2025/12/22

N2 - Laparoscopic approaches to surgery have become widely adopted, due to several advantages over open surgery. However, as an unavoidable consequence of the nature of the laparoscopic setup, skills are more difficult to acquire and sufficient quality feedback on simulator training is costly to procure. We propose a real-time algorithm to track instrument trajectories in 3D within traditional physical laparoscopic simulators. This enables feedback and performance assessment via machine learning techniques. The algorithm has been characterised using a robotically controlled laparoscopic setup as a ground truth.

AB - Laparoscopic approaches to surgery have become widely adopted, due to several advantages over open surgery. However, as an unavoidable consequence of the nature of the laparoscopic setup, skills are more difficult to acquire and sufficient quality feedback on simulator training is costly to procure. We propose a real-time algorithm to track instrument trajectories in 3D within traditional physical laparoscopic simulators. This enables feedback and performance assessment via machine learning techniques. The algorithm has been characterised using a robotically controlled laparoscopic setup as a ground truth.

U2 - 10.1109/icarm65671.2025.11293476

DO - 10.1109/icarm65671.2025.11293476

M3 - Conference contribution

BT - 2025 IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)

PB - IEEE

T2 - 10th IEEE International Conference on Advanced Robotics and Mechatronics 2025

Y2 - 1 August 2025 through 3 August 2025

ER -

Real-time Tracking of Instrument Motion to Enable Machine Learning Approaches to Laparoscopy Training

Abstract

Conference

Access to Document

Fingerprint

Cite this