Dehlela Shabir
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Dehlela Shabir has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Nikhil V. Navkar
Department of Surgery, Hamad Medical Corporation, Doha, Qatar
4 shared publications - Julien Abinahed
Technology Innovation Unit, Hamad Medical Corporation, Doha, Qatar
3 shared publications - Zhigang Deng
Department of Computer Science, University of Houston, Houston, TX, USA
3 shared publications - Hawa Hamza
Department of Surgery, Hamad Medical Corporation, Doha, Qatar
2 shared publications - Jhasketan Padhan
Department of Surgery, Hamad Medical Corporation, Doha, Qatar
2 shared publications - Khaled Shaban
Qatar University
2 shared publications
Latest Publications
- Open Access
- 23 October 2026
- Published by Elsevier in Machine Learning with Applications
- Vol. 26
Vision-based markerless surgical tool tracking in open surgery: A scoping review
- 23 October 2026
- Published by Elsevier in Machine Learning with Applications
- Vol. 26
Open AccessVision-based markerless surgical tool tracking in open surgery: A scoping review
- Open Access
- 4 March 2026
- Published by JMIR Publications Inc. in JMIR Human Factors
- Vol. 13, e86566
Dynamic Augmented Reality Cues for Telementoring in Laparoscopic Surgery: Usability Study
Background: Surgical telementoring enables a remote expert surgeon (mentor) to guide an operating surgeon (mentee) during surgery and facilitates the transfer of surgical skills. However, commonly used audio and static visual cues are inadequate to demonstrate complex tool-tissue interactions. To overcome this limitation, dynamic augmented reality (AR)–based visual cues are overlayed on the operative field to demonstrate precise instrument movements. Objective: The objective of this work was to evaluate dynamic AR cues and identify the most suitable cue that effectively demonstrates the required motions of surgical instruments during laparoscopy. Methods: A user study was conducted in a simulated environment among mentor-mentee pairs using 3 dynamic AR cues (hand gestures, a 3D pointer, and a virtual tool). The task assessed how closely the mentee was able to follow the mentor. The outcomes measured were (1) dynamic time warping distance, representing the closeness of the paths followed; (2) angular error in tooltip orientation; and (3) the NASA Task Load Index, assessing cognitive workload during telementoring. Results: Telementoring using the virtual tool resulted in a reduced dynamic time warping distance compared with hand gestures ( P <.01 for 12/13, 92.3% of the trials) and the 3D pointer ( P <.05 for 11/13, 84.6% of the trials). Lower orientation error was also noted while using the virtual tool as compared with hand gestures ( P <.05 for 5 of 6 poses). There were no substantial differences in the NASA Task Load Index scores. Conclusions: Use of a virtual tool (as a dynamic AR cue) enabled the mentee to follow the mentor’s instructions with fewer errors than both hand gestures and a 3D pointer without increasing cognitive workload. Further research is needed to assess the clinical effectiveness of the virtual tool during live surgery.
- 4 March 2026
- Published by JMIR Publications Inc. in JMIR Human Factors
- Vol. 13, e86566
Open AccessDynamic Augmented Reality Cues for Telementoring in Laparoscopic Surgery: Usability Study
Background: Surgical telementoring enables a remote expert surgeon (mentor) to guide an operating surgeon (mentee) during surgery and facilitates the transfer of surgical skills. However, commonly used audio and static visual cues are inadequate to demonstrate complex tool-tissue interactions. To overcome this limitation, dynamic augmented reality (AR)–based visual cues are overlayed on the operative field to demonstrate precise instrument movements. Objective: The objective of this work was to evaluate dynamic AR cues and identify the most suitable cue that effectively demonstrates the required motions of surgical instruments during laparoscopy. Methods: A user study was conducted in a simulated environment among mentor-mentee pairs using 3 dynamic AR cues (hand gestures, a 3D pointer, and a virtual tool). The task assessed how closely the mentee was able to follow the mentor. The outcomes measured were (1) dynamic time warping distance, representing the closeness of the paths followed; (2) angular error in tooltip orientation; and (3) the NASA Task Load Index, assessing cognitive workload during telementoring. Results: Telementoring using the virtual tool resulted in a reduced dynamic time warping distance compared with hand gestures ( P <.01 for 12/13, 92.3% of the trials) and the 3D pointer ( P <.05 for 11/13, 84.6% of the trials). Lower orientation error was also noted while using the virtual tool as compared with hand gestures ( P <.05 for 5 of 6 poses). There were no substantial differences in the NASA Task Load Index scores. Conclusions: Use of a virtual tool (as a dynamic AR cue) enabled the mentee to follow the mentor’s instructions with fewer errors than both hand gestures and a 3D pointer without increasing cognitive workload. Further research is needed to assess the clinical effectiveness of the virtual tool during live surgery.
- Open Access
- 24 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 216646-216662
A Real-Time Markerless Approach for 3D Pose Estimation of Laparoscopic Instruments in Surgical Training Simulators
Box trainers are widely used for laparoscopic skills training due to their ease of setup and realistic tool-tissue interaction. Automating skill assessment on these systems can enhance training by reducing analysis time and eliminating subjective evaluations. Achieving this requires accurate real-time estimation of the instrument tooltip’s three-dimensional (3D) pose. However, existing methods either depend on additional sensing hardware, which introduces ergonomic challenges, or lack real-time capabilities. A modular architecture has been developed for estimation of an instrument’s tooltip pose from video frames acquired during laparoscopic training. Our method incorporates a GPU-based algorithm that takes input from a customized deep learning model (YOLOv7), detects the edges of the surgical instrument, and computes tooltip’s 3D pose. Additionally, an automated 3D-pose dataset generation technique is introduced to generate datasets for training the model. Our method was evaluated using a vision-based tracking experiment, achieving 3D positional errors of 0.72 mm, 0.88 mm, and 3.81 mm over the X, Y, and Z-axes, and an orientation error of 5.65° with an end-to-end latency of 144 ms and sustained real-time throughput of 11.4 frames per second (fps). The proposed method enables accurate estimation of laparoscopic instrument tooltip poses without the need for additional hardware or markers. This work represents a significant step toward automated skill assessment in laparoscopic training using box trainers.
- 24 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 216646-216662
Open AccessA Real-Time Markerless Approach for 3D Pose Estimation of Laparoscopic Instruments in Surgical Training Simulators
Box trainers are widely used for laparoscopic skills training due to their ease of setup and realistic tool-tissue interaction. Automating skill assessment on these systems can enhance training by reducing analysis time and eliminating subjective evaluations. Achieving this requires accurate real-time estimation of the instrument tooltip’s three-dimensional (3D) pose. However, existing methods either depend on additional sensing hardware, which introduces ergonomic challenges, or lack real-time capabilities. A modular architecture has been developed for estimation of an instrument’s tooltip pose from video frames acquired during laparoscopic training. Our method incorporates a GPU-based algorithm that takes input from a customized deep learning model (YOLOv7), detects the edges of the surgical instrument, and computes tooltip’s 3D pose. Additionally, an automated 3D-pose dataset generation technique is introduced to generate datasets for training the model. Our method was evaluated using a vision-based tracking experiment, achieving 3D positional errors of 0.72 mm, 0.88 mm, and 3.81 mm over the X, Y, and Z-axes, and an orientation error of 5.65° with an end-to-end latency of 144 ms and sustained real-time throughput of 11.4 frames per second (fps). The proposed method enables accurate estimation of laparoscopic instrument tooltip poses without the need for additional hardware or markers. This work represents a significant step toward automated skill assessment in laparoscopic training using box trainers.
- Open Access
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 180-186
Deep Learning-Based Gesture Recognition for Enhanced Laparoscopic Suturing Skill Evaluation and Training
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 180-186
Open AccessDeep Learning-Based Gesture Recognition for Enhanced Laparoscopic Suturing Skill Evaluation and Training