Related Experiment Video
Updated: Apr 29, 2026
04:33
Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver
Published on: August 21, 2018
13.5K
Feasibility of real-time workflow segmentation for tracked needle interventions
IEEE Transactions on Bio-Medical Engineering
|May 22, 2014
Summary
A new algorithm accurately segments needle interventions during medical training. This computer-assisted system enhances training efficacy by providing automated feedback for procedures like epidural injections and lumbar punctures.
Area of Science:
- Medical education technology
- Medical simulation
- Interventional procedures
Background:
- Computer-assisted training systems improve medical education.
- Automatic feedback in these systems relies on workflow segmentation.
- Accurate workflow segmentation is crucial for needle-based interventions.
Purpose of the Study:
- To develop a workflow segmentation algorithm for needle interventions.
- To utilize needle tracking data for automated task determination.
- To enhance computer-assisted training systems with real-time feedback.
Main Methods:
- Collected needle tracking data from ultrasound-guided epidural injections and lumbar punctures.
- Developed a workflow segmentation algorithm using this data.
- Tested the algorithm in a simulated real-time scenario.
- Compared algorithm output to ground-truth segmentations by blinded observers.
Main Results:
- The algorithm achieved 93% overall accuracy in workflow segmentation.
- Specific accuracies included 81% for epidural procedures and 82% for lumbar punctures.
- The algorithm's accuracy surpassed manual segmentation consistency (84%).
- A medium effect size (0.5) was calculated using Cohen's d statistic.
Conclusions:
- The developed algorithm is highly accurate for segmenting needle-based procedures.
- It can augment expert observers in training systems without significant performance decrease.
- The algorithm is feasible for implementation in computer-assisted needle placement training.
