Interrater reliability of deep brain stimulation electrode localizations

Roxanne Lofredi; Cem-Georg Auernig; Siobhan Ewert; Friederike Irmen; Leon A Steiner; Ute Scheller; Bernadette C M van Wijk; Simon Oxenford; Andrea A Kühn; Andreas Horn

doi:10.1016/j.neuroimage.2022.119552

Interrater reliability of deep brain stimulation electrode localizations

Roxanne Lofredi, Cem-Georg Auernig, Siobhan Ewert, Friederike Irmen, Leon A Steiner, Ute Scheller, Bernadette C M van Wijk, Simon Oxenford, Andrea A Kühn, Andreas Horn

Research output: Contribution to Journal › Article › Academic › peer-review

Abstract

Lead-DBS is an open-source, semi-automatized and widely applied software tool facilitating precise localization of deep brain stimulation electrodes both in native as well as in standardized stereotactic space. While automatized preprocessing steps within the toolbox have been tested and validated in previous studies, the interrater reliability in manual refinements of electrode localizations using the tool has not been objectified so far. Here, we investigate the variance introduced in this processing step by different raters when localizing electrodes based on postoperative CT or MRI. Furthermore, we compare the performance of novel trainees that received a structured training and more experienced raters with an expert user. We show that all users yield similar results with an average difference in localizations ranging between 0.52-0.75 mm with 0.07-0.12 mm increases in variability when using postoperative MRI and following normalization to standard space. Our findings may pave the way toward formal training for using Lead-DBS and demonstrate its reliability and ease-of-use for imaging research in the field of deep brain stimulation.

Original language	English
Article number	119552
Pages (from-to)	119552
Journal	NeuroImage
Volume	262
DOIs	https://doi.org/10.1016/j.neuroimage.2022.119552
Publication status	Published - 15 Nov 2022

Bibliographical note

Funding

Roxanne Lofredi and Leon A. Steiner are participants in the BIH Charité (Junior) Clinician Scientist Program funded by the Charité – Universitätsmedizin Berlin, and the Berlin Institute of Health at Charité (BIH). The work was supported by Deutsche Forschungsgemeinschaft (Project-ID 424,778,381 – TRR 295 Retune). This is an EU Joint Program – Neurodegenerative Disease (JPND) project. The project is supported through the following funding organisations under the aegis of JPND - www.jpnd.eu (A.H.: the Deutsches Zentrum für Luft- und Raumfahrt - Germany; B.C.M.v.W.: the Netherlands organisation for Health Research and Development (ZonMw) - The Netherlands). A.H. was additionally supported by the German Research Foundation (Deutsche Forschungsgemeinschaft, Emmy Noether Stipend 410,169,619 and 424,778,381 – TRR 295), the Foundation for OCD research (FFOR) and the NIH (2R01 MH113929). Roxanne Lofredi and Leon A. Steiner are participants in the BIH Charité (Junior) Clinician Scientist Program funded by the Charité – Universitätsmedizin Berlin, and the Berlin Institute of Health at Charité (BIH). The work was supported by Deutsche Forschungsgemeinschaft (Project-ID 424,778,381 – TRR 295 Retune ). This is an EU Joint Program – Neurodegenerative Disease (JPND) project. The project is supported through the following funding organisations under the aegis of JPND - www.jpnd.eu (A.H.: the Deutsches Zentrum für Luft- und Raumfahrt - Germany; B.C.M.v.W.: the Netherlands organisation for Health Research and Development (ZonMw) - The Netherlands). A.H. was additionally supported by the German Research Foundation (Deutsche Forschungsgemeinschaft, Emmy Noether Stipend 410,169,619 and 424,778,381 – TRR 295 ), the Foundation for OCD research (FFOR) and the NIH (2R01 MH113929).

Funders	Funder number
Berlin Institute of Health at Charité
Foundation for OCD research
Neurodegenerative Disease
National Institutes of Health
National Institute of Mental Health	R01MH113929
EU Joint Programme – Neurodegenerative Disease Research
Deutsche Forschungsgemeinschaft	424,778,381 – TRR 295
ZonMw
Charité – Universitätsmedizin Berlin
Deutsches Zentrum für Luft- und Raumfahrt

Keywords

Deep Brain Stimulation/methods
Electrodes, Implanted
Humans
Magnetic Resonance Imaging/methods
Parkinson Disease/therapy
Reproducibility of Results
Subthalamic Nucleus/physiology

Access to Document

10.1016/j.neuroimage.2022.119552

Persistent URL (handle)

Persistent link

Cite this

@article{5abd9bb286cc44adb9811d97c997e393,

title = "Interrater reliability of deep brain stimulation electrode localizations",

abstract = "Lead-DBS is an open-source, semi-automatized and widely applied software tool facilitating precise localization of deep brain stimulation electrodes both in native as well as in standardized stereotactic space. While automatized preprocessing steps within the toolbox have been tested and validated in previous studies, the interrater reliability in manual refinements of electrode localizations using the tool has not been objectified so far. Here, we investigate the variance introduced in this processing step by different raters when localizing electrodes based on postoperative CT or MRI. Furthermore, we compare the performance of novel trainees that received a structured training and more experienced raters with an expert user. We show that all users yield similar results with an average difference in localizations ranging between 0.52-0.75 mm with 0.07-0.12 mm increases in variability when using postoperative MRI and following normalization to standard space. Our findings may pave the way toward formal training for using Lead-DBS and demonstrate its reliability and ease-of-use for imaging research in the field of deep brain stimulation.",

keywords = "Deep Brain Stimulation/methods, Electrodes, Implanted, Humans, Magnetic Resonance Imaging/methods, Parkinson Disease/therapy, Reproducibility of Results, Subthalamic Nucleus/physiology",

author = "Roxanne Lofredi and Cem-Georg Auernig and Siobhan Ewert and Friederike Irmen and Steiner, {Leon A} and Ute Scheller and {van Wijk}, {Bernadette C M} and Simon Oxenford and K{\"u}hn, {Andrea A} and Andreas Horn",

note = "Copyright {\textcopyright} 2022. Published by Elsevier Inc.",

year = "2022",

month = nov,

day = "15",

doi = "10.1016/j.neuroimage.2022.119552",

language = "English",

volume = "262",

pages = "119552",

journal = "NeuroImage",

issn = "1053-8119",

publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Interrater reliability of deep brain stimulation electrode localizations

AU - Lofredi, Roxanne

AU - Auernig, Cem-Georg

AU - Ewert, Siobhan

AU - Irmen, Friederike

AU - Steiner, Leon A

AU - Scheller, Ute

AU - van Wijk, Bernadette C M

AU - Oxenford, Simon

AU - Kühn, Andrea A

AU - Horn, Andreas

PY - 2022/11/15

Y1 - 2022/11/15

N2 - Lead-DBS is an open-source, semi-automatized and widely applied software tool facilitating precise localization of deep brain stimulation electrodes both in native as well as in standardized stereotactic space. While automatized preprocessing steps within the toolbox have been tested and validated in previous studies, the interrater reliability in manual refinements of electrode localizations using the tool has not been objectified so far. Here, we investigate the variance introduced in this processing step by different raters when localizing electrodes based on postoperative CT or MRI. Furthermore, we compare the performance of novel trainees that received a structured training and more experienced raters with an expert user. We show that all users yield similar results with an average difference in localizations ranging between 0.52-0.75 mm with 0.07-0.12 mm increases in variability when using postoperative MRI and following normalization to standard space. Our findings may pave the way toward formal training for using Lead-DBS and demonstrate its reliability and ease-of-use for imaging research in the field of deep brain stimulation.

AB - Lead-DBS is an open-source, semi-automatized and widely applied software tool facilitating precise localization of deep brain stimulation electrodes both in native as well as in standardized stereotactic space. While automatized preprocessing steps within the toolbox have been tested and validated in previous studies, the interrater reliability in manual refinements of electrode localizations using the tool has not been objectified so far. Here, we investigate the variance introduced in this processing step by different raters when localizing electrodes based on postoperative CT or MRI. Furthermore, we compare the performance of novel trainees that received a structured training and more experienced raters with an expert user. We show that all users yield similar results with an average difference in localizations ranging between 0.52-0.75 mm with 0.07-0.12 mm increases in variability when using postoperative MRI and following normalization to standard space. Our findings may pave the way toward formal training for using Lead-DBS and demonstrate its reliability and ease-of-use for imaging research in the field of deep brain stimulation.

KW - Deep Brain Stimulation/methods

KW - Electrodes, Implanted

KW - Humans

KW - Magnetic Resonance Imaging/methods

KW - Parkinson Disease/therapy

KW - Reproducibility of Results

KW - Subthalamic Nucleus/physiology

U2 - 10.1016/j.neuroimage.2022.119552

DO - 10.1016/j.neuroimage.2022.119552

M3 - Article

C2 - 35981644

SN - 1053-8119

VL - 262

SP - 119552

JO - NeuroImage

JF - NeuroImage

M1 - 119552

ER -

Interrater reliability of deep brain stimulation electrode localizations

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Persistent URL (handle)

Fingerprint

Cite this