Clinical Translation of the LevelCheck Decision Support Algorithm for Target Localization in Spine Surgery

Amir Manbachi, Tharindu De Silva, Ali Uneri, Matthew Jacobson, Joseph Goerres, Michael Ketcha, Runze Han, Nafi Aygun, David Thompson, Xiaobu Ye, Sebastian Vogt, Gerhard Kleinszig, Camilo Molina, Rajiv Iyer, Tomas Garzon-Muvdi, Michael R. Raber, Mari Groves, Jean Paul Wolinsky, Jeffrey H. Siewerdsen

Research output: Contribution to journalArticle

Abstract

Recent work has yielded a method for automatic labeling of vertebrae in intraoperative radiographs as an assistant to manual level counting. The method, called LevelCheck, previously demonstrated promise in phantom studies and retrospective studies. This study aims to: (#1) Analyze the effect of LevelCheck on accuracy and confidence of localization in two modes: (a) Independent Check (labels displayed after the surgeon’s decision) and (b) Active Assistant (labels presented before the surgeon’s decision). (#2) Assess the feasibility and utility of LevelCheck in the operating room. Two studies were conducted: a laboratory study investigating these two workflow implementations in a simulated operating environment with 5 surgeons, reviewing 62 cases selected from a dataset of radiographs exhibiting a challenge to vertebral localization; and a clinical study involving 20 patients undergoing spine surgery. In Study #1, the median localization error without assistance was 30.4% (IQR = 5.2%) due to the challenging nature of the cases. LevelCheck reduced the median error to 2.4% for both the Independent Check and Active Assistant modes (p < 0.01). Surgeons found LevelCheck to increase confidence in 91% of cases. Study #2 demonstrated accuracy in all cases. The algorithm runtime varied from 17 to 72 s in its current implementation. The algorithm was shown to be feasible, accurate, and to improve confidence during surgery.

Original languageEnglish (US)
JournalAnnals of Biomedical Engineering
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Surgery
Labels
Operating rooms
Labeling

Keywords

  • Clinical translation
  • Image-guided surgery
  • Intraoperative imaging
  • LevelCheck
  • Spine surgery
  • Surgical workflow

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Clinical Translation of the LevelCheck Decision Support Algorithm for Target Localization in Spine Surgery. / Manbachi, Amir; De Silva, Tharindu; Uneri, Ali; Jacobson, Matthew; Goerres, Joseph; Ketcha, Michael; Han, Runze; Aygun, Nafi; Thompson, David; Ye, Xiaobu; Vogt, Sebastian; Kleinszig, Gerhard; Molina, Camilo; Iyer, Rajiv; Garzon-Muvdi, Tomas; Raber, Michael R.; Groves, Mari; Wolinsky, Jean Paul; Siewerdsen, Jeffrey H.

In: Annals of Biomedical Engineering, 01.01.2018.

Research output: Contribution to journalArticle

Manbachi, Amir ; De Silva, Tharindu ; Uneri, Ali ; Jacobson, Matthew ; Goerres, Joseph ; Ketcha, Michael ; Han, Runze ; Aygun, Nafi ; Thompson, David ; Ye, Xiaobu ; Vogt, Sebastian ; Kleinszig, Gerhard ; Molina, Camilo ; Iyer, Rajiv ; Garzon-Muvdi, Tomas ; Raber, Michael R. ; Groves, Mari ; Wolinsky, Jean Paul ; Siewerdsen, Jeffrey H. / Clinical Translation of the LevelCheck Decision Support Algorithm for Target Localization in Spine Surgery. In: Annals of Biomedical Engineering. 2018.
@article{5f931305798a453a932ad1b51da0ab41,
title = "Clinical Translation of the LevelCheck Decision Support Algorithm for Target Localization in Spine Surgery",
abstract = "Recent work has yielded a method for automatic labeling of vertebrae in intraoperative radiographs as an assistant to manual level counting. The method, called LevelCheck, previously demonstrated promise in phantom studies and retrospective studies. This study aims to: (#1) Analyze the effect of LevelCheck on accuracy and confidence of localization in two modes: (a) Independent Check (labels displayed after the surgeon’s decision) and (b) Active Assistant (labels presented before the surgeon’s decision). (#2) Assess the feasibility and utility of LevelCheck in the operating room. Two studies were conducted: a laboratory study investigating these two workflow implementations in a simulated operating environment with 5 surgeons, reviewing 62 cases selected from a dataset of radiographs exhibiting a challenge to vertebral localization; and a clinical study involving 20 patients undergoing spine surgery. In Study #1, the median localization error without assistance was 30.4{\%} (IQR = 5.2{\%}) due to the challenging nature of the cases. LevelCheck reduced the median error to 2.4{\%} for both the Independent Check and Active Assistant modes (p < 0.01). Surgeons found LevelCheck to increase confidence in 91{\%} of cases. Study #2 demonstrated accuracy in all cases. The algorithm runtime varied from 17 to 72 s in its current implementation. The algorithm was shown to be feasible, accurate, and to improve confidence during surgery.",
keywords = "Clinical translation, Image-guided surgery, Intraoperative imaging, LevelCheck, Spine surgery, Surgical workflow",
author = "Amir Manbachi and {De Silva}, Tharindu and Ali Uneri and Matthew Jacobson and Joseph Goerres and Michael Ketcha and Runze Han and Nafi Aygun and David Thompson and Xiaobu Ye and Sebastian Vogt and Gerhard Kleinszig and Camilo Molina and Rajiv Iyer and Tomas Garzon-Muvdi and Raber, {Michael R.} and Mari Groves and Wolinsky, {Jean Paul} and Siewerdsen, {Jeffrey H.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/s10439-018-2099-2",
language = "English (US)",
journal = "Annals of Biomedical Engineering",
issn = "0090-6964",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Clinical Translation of the LevelCheck Decision Support Algorithm for Target Localization in Spine Surgery

AU - Manbachi, Amir

AU - De Silva, Tharindu

AU - Uneri, Ali

AU - Jacobson, Matthew

AU - Goerres, Joseph

AU - Ketcha, Michael

AU - Han, Runze

AU - Aygun, Nafi

AU - Thompson, David

AU - Ye, Xiaobu

AU - Vogt, Sebastian

AU - Kleinszig, Gerhard

AU - Molina, Camilo

AU - Iyer, Rajiv

AU - Garzon-Muvdi, Tomas

AU - Raber, Michael R.

AU - Groves, Mari

AU - Wolinsky, Jean Paul

AU - Siewerdsen, Jeffrey H.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Recent work has yielded a method for automatic labeling of vertebrae in intraoperative radiographs as an assistant to manual level counting. The method, called LevelCheck, previously demonstrated promise in phantom studies and retrospective studies. This study aims to: (#1) Analyze the effect of LevelCheck on accuracy and confidence of localization in two modes: (a) Independent Check (labels displayed after the surgeon’s decision) and (b) Active Assistant (labels presented before the surgeon’s decision). (#2) Assess the feasibility and utility of LevelCheck in the operating room. Two studies were conducted: a laboratory study investigating these two workflow implementations in a simulated operating environment with 5 surgeons, reviewing 62 cases selected from a dataset of radiographs exhibiting a challenge to vertebral localization; and a clinical study involving 20 patients undergoing spine surgery. In Study #1, the median localization error without assistance was 30.4% (IQR = 5.2%) due to the challenging nature of the cases. LevelCheck reduced the median error to 2.4% for both the Independent Check and Active Assistant modes (p < 0.01). Surgeons found LevelCheck to increase confidence in 91% of cases. Study #2 demonstrated accuracy in all cases. The algorithm runtime varied from 17 to 72 s in its current implementation. The algorithm was shown to be feasible, accurate, and to improve confidence during surgery.

AB - Recent work has yielded a method for automatic labeling of vertebrae in intraoperative radiographs as an assistant to manual level counting. The method, called LevelCheck, previously demonstrated promise in phantom studies and retrospective studies. This study aims to: (#1) Analyze the effect of LevelCheck on accuracy and confidence of localization in two modes: (a) Independent Check (labels displayed after the surgeon’s decision) and (b) Active Assistant (labels presented before the surgeon’s decision). (#2) Assess the feasibility and utility of LevelCheck in the operating room. Two studies were conducted: a laboratory study investigating these two workflow implementations in a simulated operating environment with 5 surgeons, reviewing 62 cases selected from a dataset of radiographs exhibiting a challenge to vertebral localization; and a clinical study involving 20 patients undergoing spine surgery. In Study #1, the median localization error without assistance was 30.4% (IQR = 5.2%) due to the challenging nature of the cases. LevelCheck reduced the median error to 2.4% for both the Independent Check and Active Assistant modes (p < 0.01). Surgeons found LevelCheck to increase confidence in 91% of cases. Study #2 demonstrated accuracy in all cases. The algorithm runtime varied from 17 to 72 s in its current implementation. The algorithm was shown to be feasible, accurate, and to improve confidence during surgery.

KW - Clinical translation

KW - Image-guided surgery

KW - Intraoperative imaging

KW - LevelCheck

KW - Spine surgery

KW - Surgical workflow

UR - http://www.scopus.com/inward/record.url?scp=85050761430&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050761430&partnerID=8YFLogxK

U2 - 10.1007/s10439-018-2099-2

DO - 10.1007/s10439-018-2099-2

M3 - Article

C2 - 30051244

AN - SCOPUS:85050761430

JO - Annals of Biomedical Engineering

JF - Annals of Biomedical Engineering

SN - 0090-6964

ER -