Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks

Martin Pfister; Kornelia Schützenberger; Ulrike Pfeiffenberger; Alina Messner; Z. H.E. Chen; Valentin Aranha Dos Santos; Stefan Puchner; Gerhard Garhöfer; Leopold Schmetterer; Martin Gröschl; René M. Werkmeister

doi:10.1364/BOE.10.001315

Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks

Martin Pfister, Kornelia Schützenberger, Ulrike Pfeiffenberger, Alina Messner, Z. H.E. Chen, Valentin Aranha Dos Santos, Stefan Puchner, Gerhard Garhöfer, Leopold Schmetterer, Martin Gröschl, René M. Werkmeister^*

^*Corresponding author for this work

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

16 Citations (Scopus)

Abstract

We present a system for automatic determination of the intradermal volume of hydrogels based on optical coherence tomography (OCT) and deep learning. Volumetric image data was acquired using a custom-built OCT prototype that employs an akinetic swept laser at ∼1310 nm with a bandwidth of 87 nm, providing an axial resolution of ∼6.5 µm in tissue. Three-dimensional data sets of a 10 mm × 10 mm skin patch comprising the intradermal filler and the surrounding tissue were acquired. A convolutional neural network using a u-net-like architecture was trained from slices of 100 OCT volume data sets where the dermal filler volume was manually annotated. Using six-fold cross-validation, a mean accuracy of 0.9938 and a Jaccard similarity coefficient of 0.879 were achieved.

Original language	English
Article number	#347728
Pages (from-to)	1315-1328
Number of pages	14
Journal	Biomedical Optics Express
Volume	10
Issue number	3
DOIs	https://doi.org/10.1364/BOE.10.001315
Publication status	Published - Mar 1 2019
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement.

ASJC Scopus Subject Areas

Biotechnology
Atomic and Molecular Physics, and Optics

Access to Document

10.1364/BOE.10.001315

Cite this

Pfister, M., Schützenberger, K., Pfeiffenberger, U., Messner, A., Chen, Z. H. E., Santos, V. A. D., Puchner, S., Garhöfer, G., Schmetterer, L., Gröschl, M., & Werkmeister, R. M. (2019). Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks. Biomedical Optics Express, 10(3), 1315-1328. Article #347728. https://doi.org/10.1364/BOE.10.001315

@article{382ceae2b79745b7aaab2d17b5892af2,

title = "Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks",

abstract = "We present a system for automatic determination of the intradermal volume of hydrogels based on optical coherence tomography (OCT) and deep learning. Volumetric image data was acquired using a custom-built OCT prototype that employs an akinetic swept laser at ∼1310 nm with a bandwidth of 87 nm, providing an axial resolution of ∼6.5 µm in tissue. Three-dimensional data sets of a 10 mm × 10 mm skin patch comprising the intradermal filler and the surrounding tissue were acquired. A convolutional neural network using a u-net-like architecture was trained from slices of 100 OCT volume data sets where the dermal filler volume was manually annotated. Using six-fold cross-validation, a mean accuracy of 0.9938 and a Jaccard similarity coefficient of 0.879 were achieved.",

author = "Martin Pfister and Kornelia Sch{\"u}tzenberger and Ulrike Pfeiffenberger and Alina Messner and Chen, {Z. H.E.} and Santos, {Valentin Aranha Dos} and Stefan Puchner and Gerhard Garh{\"o}fer and Leopold Schmetterer and Martin Gr{\"o}schl and Werkmeister, {Ren{\'e} M.}",

note = "Publisher Copyright: {\textcopyright} 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement.",

year = "2019",

month = mar,

day = "1",

doi = "10.1364/BOE.10.001315",

language = "English",

volume = "10",

pages = "1315--1328",

journal = "Biomedical Optics Express",

issn = "2156-7085",

publisher = "The Optical Society",

number = "3",

}

Pfister, M, Schützenberger, K, Pfeiffenberger, U, Messner, A, Chen, ZHE, Santos, VAD, Puchner, S, Garhöfer, G, Schmetterer, L, Gröschl, M & Werkmeister, RM 2019, 'Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks', Biomedical Optics Express, vol. 10, no. 3, #347728, pp. 1315-1328. https://doi.org/10.1364/BOE.10.001315

TY - JOUR

T1 - Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks

AU - Pfister, Martin

AU - Schützenberger, Kornelia

AU - Pfeiffenberger, Ulrike

AU - Messner, Alina

AU - Chen, Z. H.E.

AU - Santos, Valentin Aranha Dos

AU - Puchner, Stefan

AU - Garhöfer, Gerhard

AU - Schmetterer, Leopold

AU - Gröschl, Martin

AU - Werkmeister, René M.

PY - 2019/3/1

Y1 - 2019/3/1

N2 - We present a system for automatic determination of the intradermal volume of hydrogels based on optical coherence tomography (OCT) and deep learning. Volumetric image data was acquired using a custom-built OCT prototype that employs an akinetic swept laser at ∼1310 nm with a bandwidth of 87 nm, providing an axial resolution of ∼6.5 µm in tissue. Three-dimensional data sets of a 10 mm × 10 mm skin patch comprising the intradermal filler and the surrounding tissue were acquired. A convolutional neural network using a u-net-like architecture was trained from slices of 100 OCT volume data sets where the dermal filler volume was manually annotated. Using six-fold cross-validation, a mean accuracy of 0.9938 and a Jaccard similarity coefficient of 0.879 were achieved.

AB - We present a system for automatic determination of the intradermal volume of hydrogels based on optical coherence tomography (OCT) and deep learning. Volumetric image data was acquired using a custom-built OCT prototype that employs an akinetic swept laser at ∼1310 nm with a bandwidth of 87 nm, providing an axial resolution of ∼6.5 µm in tissue. Three-dimensional data sets of a 10 mm × 10 mm skin patch comprising the intradermal filler and the surrounding tissue were acquired. A convolutional neural network using a u-net-like architecture was trained from slices of 100 OCT volume data sets where the dermal filler volume was manually annotated. Using six-fold cross-validation, a mean accuracy of 0.9938 and a Jaccard similarity coefficient of 0.879 were achieved.

UR - http://www.scopus.com/inward/record.url?scp=85063931966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063931966&partnerID=8YFLogxK

U2 - 10.1364/BOE.10.001315

DO - 10.1364/BOE.10.001315

M3 - Article

AN - SCOPUS:85063931966

SN - 2156-7085

VL - 10

SP - 1315

EP - 1328

JO - Biomedical Optics Express

JF - Biomedical Optics Express

IS - 3

M1 - #347728

ER -

Automated segmentation of dermal fillers in OCT images of mice using convolutional neural networks

Abstract

Bibliographical note

ASJC Scopus Subject Areas

Access to Document

Other files and links

Fingerprint

Cite this