Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT

Marc B. Muijzer; Friso G. Heslinga; Floor Couwenberg; Herke Jan Noordmans; Abdelkarim Oahalou; Josien P.W. Pluim; Mitko Veta; Robert P.L. Wisse

doi:10.1364/BOE.446519

Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT

Marc B. Muijzer (Corresponding author)
, Friso G. Heslinga
, Floor Couwenberg
, Herke Jan Noordmans
, Abdelkarim Oahalou
, Josien P.W. Pluim
, Mitko Veta
, Robert P.L. Wisse

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

8 Citations (Scopus)

204 Downloads (Pure)

Abstract

Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86%. In comparison, two corneal specialists correctly identified graft orientation in 90% and 91% of the iOCT-frames.

Original language	English
Pages (from-to)	2683-2694
Number of pages	12
Journal	Biomedical Optics Express
Volume	13
Issue number	5
DOIs	https://doi.org/10.1364/BOE.446519
Publication status	Published - 1 May 2022

Bibliographical note

Funding Information:
F.P. Fischer stichting (FS 2020-1, UZ 2016-14); Carl Zeiss Meditec AG; Philips; Nederlandse Organisatie voor Wetenschappelijk Onderzoek. The authors are grateful to Marijn Hiep, Suzanne de Vette, and Inge van den Berg for their assistance with the development of the DMEK orientation model and conceptualization of a user interface.

Funding

F.P. Fischer stichting (FS 2020-1, UZ 2016-14); Carl Zeiss Meditec AG; Philips; Nederlandse Organisatie voor Wetenschappelijk Onderzoek. The authors are grateful to Marijn Hiep, Suzanne de Vette, and Inge van den Berg for their assistance with the development of the DMEK orientation model and conceptualization of a user interface.

Access to Document

10.1364/BOE.446519

pdfFinal published version, 3.53 MBLicence: CC BY

Cite this

@article{47b60e69a00e4b9085a67282d60e6c88,

title = "Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT",

abstract = "Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86\%. In comparison, two corneal specialists correctly identified graft orientation in 90\% and 91\% of the iOCT-frames.",

author = "Muijzer, \{Marc B.\} and Heslinga, \{Friso G.\} and Floor Couwenberg and Noordmans, \{Herke Jan\} and Abdelkarim Oahalou and Pluim, \{Josien P.W.\} and Mitko Veta and Wisse, \{Robert P.L.\}",

note = "Funding Information: F.P. Fischer stichting (FS 2020-1, UZ 2016-14); Carl Zeiss Meditec AG; Philips; Nederlandse Organisatie voor Wetenschappelijk Onderzoek. The authors are grateful to Marijn Hiep, Suzanne de Vette, and Inge van den Berg for their assistance with the development of the DMEK orientation model and conceptualization of a user interface. ",

year = "2022",

month = may,

day = "1",

doi = "10.1364/BOE.446519",

language = "English",

volume = "13",

pages = "2683--2694",

journal = "Biomedical Optics Express",

issn = "2156-7085",

publisher = "Optica Publishing Group",

number = "5",

}

TY - JOUR

T1 - Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT

AU - Muijzer, Marc B.

AU - Heslinga, Friso G.

AU - Couwenberg, Floor

AU - Noordmans, Herke Jan

AU - Oahalou, Abdelkarim

AU - Pluim, Josien P.W.

AU - Veta, Mitko

AU - Wisse, Robert P.L.

N1 - Funding Information: F.P. Fischer stichting (FS 2020-1, UZ 2016-14); Carl Zeiss Meditec AG; Philips; Nederlandse Organisatie voor Wetenschappelijk Onderzoek. The authors are grateful to Marijn Hiep, Suzanne de Vette, and Inge van den Berg for their assistance with the development of the DMEK orientation model and conceptualization of a user interface.

PY - 2022/5/1

Y1 - 2022/5/1

N2 - Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86%. In comparison, two corneal specialists correctly identified graft orientation in 90% and 91% of the iOCT-frames.

AB - Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86%. In comparison, two corneal specialists correctly identified graft orientation in 90% and 91% of the iOCT-frames.

UR - https://www.scopus.com/pages/publications/85128638868

U2 - 10.1364/BOE.446519

DO - 10.1364/BOE.446519

M3 - Article

C2 - 35774322

AN - SCOPUS:85128638868

SN - 2156-7085

VL - 13

SP - 2683

EP - 2694

JO - Biomedical Optics Express

JF - Biomedical Optics Express

IS - 5

ER -

Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT

Abstract

Bibliographical note

Funding

Access to Document

Other files and links

Fingerprint

Cite this