Medical instrument detection in 3-dimensional ultrasound data volumes

Research output: Contribution to journalArticleAcademicpeer-review

11 Citations (Scopus)

Abstract

Ultrasound-guided medical interventions are broadly applied in diagnostics and therapy, e.g. regional anesthesia or ablation. A guided intervention using 2D ultrasound is challenging due to the poor instrument visibility, limited field of view and the multi-fold coordination of the medical instrument and ultrasound plane. Recent 3D ultrasound transducers can improve the quality of the image-guided intervention if an automated detection of the needle is used. In this paper, we present a novel method for detecting medical instruments in 3D ultrasound data that is solely based on image processing techniques and validated on various ex-vivo and in-vivo datasets. In the proposed procedure, the physician is placing the 3D transducer at the desired position and the image processing will automatically detect the best instrument view, so that the physician can entirely focus on the intervention. Our method is based on classification of instrument voxels using volumetric structure directions and robust approximation of the primary tool axis. A novel normalization method is proposed for the shape and intensity consistency of instruments to improve the detection. Moreover, a novel 3D Gabor wavelet transformation is introduced and optimally designed for revealing the instrument voxels in the volume, while remaining generic to several medical instruments and transducer types. Experiments on diverse datasets including in-vivo data from patients show that for a given transducer and instrument type, high detection accuracies are achieved with position errors smaller than the instrument diameter in the 0.5 to 1.5 millimeter range on average.
LanguageEnglish
Pages1664-1675
Number of pages12
JournalIEEE Transactions on Medical Imaging
Volume36
Issue number8
DOIs
StatePublished - 1 Aug 2017

Fingerprint

Transducers
Ultrasonics
Physicians
Conduction Anesthesia
Needles
Image processing
Ablation
Visibility
Datasets
Therapeutics

Cite this

@article{26fd13b4bb1942cc804142512b27ef08,
title = "Medical instrument detection in 3-dimensional ultrasound data volumes",
abstract = "Ultrasound-guided medical interventions are broadly applied in diagnostics and therapy, e.g. regional anesthesia or ablation. A guided intervention using 2D ultrasound is challenging due to the poor instrument visibility, limited field of view and the multi-fold coordination of the medical instrument and ultrasound plane. Recent 3D ultrasound transducers can improve the quality of the image-guided intervention if an automated detection of the needle is used. In this paper, we present a novel method for detecting medical instruments in 3D ultrasound data that is solely based on image processing techniques and validated on various ex-vivo and in-vivo datasets. In the proposed procedure, the physician is placing the 3D transducer at the desired position and the image processing will automatically detect the best instrument view, so that the physician can entirely focus on the intervention. Our method is based on classification of instrument voxels using volumetric structure directions and robust approximation of the primary tool axis. A novel normalization method is proposed for the shape and intensity consistency of instruments to improve the detection. Moreover, a novel 3D Gabor wavelet transformation is introduced and optimally designed for revealing the instrument voxels in the volume, while remaining generic to several medical instruments and transducer types. Experiments on diverse datasets including in-vivo data from patients show that for a given transducer and instrument type, high detection accuracies are achieved with position errors smaller than the instrument diameter in the 0.5 to 1.5 millimeter range on average.",
author = "A. Pourtaherian and H.J. Scholten and C.J. Kusters and S. Zinger and N. Mihajlovic and A.F. Kolen and F. Zou and G.C. Ng and H. Korsten and {de With}, P.H.N.",
year = "2017",
month = "8",
day = "1",
doi = "10.1109/TMI.2017.2692302",
language = "English",
volume = "36",
pages = "1664--1675",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers",
number = "8",

}

Medical instrument detection in 3-dimensional ultrasound data volumes. / Pourtaherian, A.; Scholten, H.J.; Kusters, C.J.; Zinger, S.; Mihajlovic, N.; Kolen, A.F.; Zou, F.; Ng, G.C.; Korsten, H.; de With, P.H.N.

In: IEEE Transactions on Medical Imaging, Vol. 36, No. 8, 01.08.2017, p. 1664-1675.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Medical instrument detection in 3-dimensional ultrasound data volumes

AU - Pourtaherian,A.

AU - Scholten,H.J.

AU - Kusters,C.J.

AU - Zinger,S.

AU - Mihajlovic,N.

AU - Kolen,A.F.

AU - Zou,F.

AU - Ng,G.C.

AU - Korsten,H.

AU - de With,P.H.N.

PY - 2017/8/1

Y1 - 2017/8/1

N2 - Ultrasound-guided medical interventions are broadly applied in diagnostics and therapy, e.g. regional anesthesia or ablation. A guided intervention using 2D ultrasound is challenging due to the poor instrument visibility, limited field of view and the multi-fold coordination of the medical instrument and ultrasound plane. Recent 3D ultrasound transducers can improve the quality of the image-guided intervention if an automated detection of the needle is used. In this paper, we present a novel method for detecting medical instruments in 3D ultrasound data that is solely based on image processing techniques and validated on various ex-vivo and in-vivo datasets. In the proposed procedure, the physician is placing the 3D transducer at the desired position and the image processing will automatically detect the best instrument view, so that the physician can entirely focus on the intervention. Our method is based on classification of instrument voxels using volumetric structure directions and robust approximation of the primary tool axis. A novel normalization method is proposed for the shape and intensity consistency of instruments to improve the detection. Moreover, a novel 3D Gabor wavelet transformation is introduced and optimally designed for revealing the instrument voxels in the volume, while remaining generic to several medical instruments and transducer types. Experiments on diverse datasets including in-vivo data from patients show that for a given transducer and instrument type, high detection accuracies are achieved with position errors smaller than the instrument diameter in the 0.5 to 1.5 millimeter range on average.

AB - Ultrasound-guided medical interventions are broadly applied in diagnostics and therapy, e.g. regional anesthesia or ablation. A guided intervention using 2D ultrasound is challenging due to the poor instrument visibility, limited field of view and the multi-fold coordination of the medical instrument and ultrasound plane. Recent 3D ultrasound transducers can improve the quality of the image-guided intervention if an automated detection of the needle is used. In this paper, we present a novel method for detecting medical instruments in 3D ultrasound data that is solely based on image processing techniques and validated on various ex-vivo and in-vivo datasets. In the proposed procedure, the physician is placing the 3D transducer at the desired position and the image processing will automatically detect the best instrument view, so that the physician can entirely focus on the intervention. Our method is based on classification of instrument voxels using volumetric structure directions and robust approximation of the primary tool axis. A novel normalization method is proposed for the shape and intensity consistency of instruments to improve the detection. Moreover, a novel 3D Gabor wavelet transformation is introduced and optimally designed for revealing the instrument voxels in the volume, while remaining generic to several medical instruments and transducer types. Experiments on diverse datasets including in-vivo data from patients show that for a given transducer and instrument type, high detection accuracies are achieved with position errors smaller than the instrument diameter in the 0.5 to 1.5 millimeter range on average.

U2 - 10.1109/TMI.2017.2692302

DO - 10.1109/TMI.2017.2692302

M3 - Article

VL - 36

SP - 1664

EP - 1675

JO - IEEE Transactions on Medical Imaging

T2 - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 8

ER -