ConVibNet improves real-time needle tip and angle detection in ultrasound by modeling temporal correlations with a novel intersection-and-difference loss, achieving 2.80 mm tip error and 0.75 mm gain over baselines.
In: 2023 IEEE/RSJ Inter- national Conference on Intelligent Robots and Systems (IROS), pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ConVibNet: Needle Detection during Continuous Insertion via Frequency-Inspired Features
ConVibNet improves real-time needle tip and angle detection in ultrasound by modeling temporal correlations with a novel intersection-and-difference loss, achieving 2.80 mm tip error and 0.75 mm gain over baselines.