Lumbar Puncture is a process in which a needle is inserted between two lumbar vertebrae, to access cerebrospinal fluid. While the use of ultrasound for visualization is common, it is often noisy and fails to provide a complete view of the region. This thesis advances lumbar puncture techniques by integrating motion tracked 2D ultrasound and 3D MRI imaging, for providing visualization of the interspinal spaces of the lumbar vertebrae. We create a method where point clouds derived from ultrasound scans of the lumbar spine are registered with a volumetric MRI ground truth model. This facilitates the determination of safe needle insertion zones, which are then dynamically displayed on the ultrasound screen. This approach has the potential to aid clinicians real-time during LP procedures, which may result in reduced risk of procedural complications. Our work represents starting point towards enhanced image guided lumbar puncture. In this thesis, we present a framework for localizing valid trajectories based on acquired ultrasound images by using an MRI ground truth as guidance. The proposed pipeline, upon capturing an ultrasound image, is designed to determine the closest needle puncture location specified within the MRI ground truth point cloud.

Creator

• Tamhankar, Aabha

Colaboradores

• Zhang, Haichong
• Nycz, Christopher J.
• Zheng, Yihao

Degree

• MS

Unit

• Robotics Engineering

Publisher

• Worcester Polytechnic Institute

Identifier

• etd-121267

Palabra Clave

• Anatomical Localization
• Lumbar Puncture Procedure
• Ultrasound Imaging
• Needle Insertion Guidance
• MRI

Advisor

• Nycz, Christopher J.

Committee

• Zhang, Haichong
• Zheng, Yihao

Defense date

• 2024-04-26

Year

• 2024

Date created

• 2024-04-22

Resource type

• Thesis

Source

• etd-121267

Rights statement

• In Copyright