Ultrasound imaging is commonly used to guide the placement of needles for tissue biopsies, catheter insertions, drainages, and anesthesia. However, visual artifacts make it difficult to locate the needle in cases of steep insertion angles (40-80 degrees) and depths (9 cm). Hence, there is a need for a method for real-time needle detection.
Researchers at Rutgers University developed a novel method for automatically localizing a needle and other surgical tools using 2D/3D ultrasound in real-time. By applying deep learning-based convolution neural networks, this method addresses enhancement and localization, thereby improving the ability of computer vision systems to accurately detect needles in real-time during ultrasound-guided procedures.
This software is cohesive with current 2D and 3D cart-based and hand-held ultrasound systems for facilitated integration.
Accurate needle visualization and localization during ultrasound-guided procedures like:
- Percutaneous biopsies
- Central venous cannulation
- Regional anesthesia
- Fetal medicine
- Locates moving surgical tools at steep insertion angles and depths
- Easily integrated into current ultrasound machines
- May be applied to other digital imaging techniques
- Realtime (65fps) processing speed
Intellectual Property & Development Status:
Patent pending. Available for licensing or research collaboration.
- Mwikirize C, Nosher JL, Hacihaliloglu I. Convolution neural networks for real-time needle detection and localization in 2D ultrasound. Int J CARS (2018)
- Mwikirize C, Nosher JL, Hacihaliloglu I. Signal attenuation maps for needle enhancement and localization in 2D ultrasound. Int J CARS (2018)
- Nosher J, Mwikirize C, Hacihaliloglu I. Method for realtime localization of inplane and out-of-plane needles from 2D ultrasound. Int J CARS (2019)