Model Fitting using RANSAC for Surgical Tool Localization in 3D Ultrasound Images

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F09%3A00162857" target="_blank" >RIV/68407700:21230/09:00162857 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Model Fitting using RANSAC for Surgical Tool Localization in 3D Ultrasound Images

Popis výsledku v původním jazyce

Ultrasound guidance is used for many surgical interventions like biopsy or electrode insertion. We present a method for localization of a thin surgical tool such as a biopsy needle or a micro-electrode in a three dimensional (3D) ultrasound image. The proposed method starts with thresholding and model fitting using RANSAC for robust localization of the axis. Subsequent local optimization refines its position. Finally, the tip of the tool is localized by finding an intensity drop along the axis. Two different tool models are presented, one simple and fast, the second using a learnt a priori information about voxel intensities of the tool and the background. The simulation study shows that our algorithm can localize the tool at almost real time speed even using a Matlab implementation, with accuracy better than 1 mm. In experimental comparison to several alternative localization methods our method appears to be the fastest and the most robust one. We also show results on real 3D ultrasou

Název v anglickém jazyce

Model Fitting using RANSAC for Surgical Tool Localization in 3D Ultrasound Images

Popis výsledku anglicky

Ultrasound guidance is used for many surgical interventions like biopsy or electrode insertion. We present a method for localization of a thin surgical tool such as a biopsy needle or a micro-electrode in a three dimensional (3D) ultrasound image. The proposed method starts with thresholding and model fitting using RANSAC for robust localization of the axis. Subsequent local optimization refines its position. Finally, the tip of the tool is localized by finding an intensity drop along the axis. Two different tool models are presented, one simple and fast, the second using a learnt a priori information about voxel intensities of the tool and the background. The simulation study shows that our algorithm can localize the tool at almost real time speed even using a Matlab implementation, with accuracy better than 1 mm. In experimental comparison to several alternative localization methods our method appears to be the fastest and the most robust one. We also show results on real 3D ultrasou

Druh

O - Ostatní výsledky

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2009

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů