A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer

Mathis Ersted Rasmussen; Jasper Albertus Nijkamp; Jesper Grau Eriksen; Stine Sofia Korreman

doi:10.1016/j.phro.2023.100426

A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer

Mathis Ersted Rasmussen, Jasper Albertus Nijkamp, Jesper Grau Eriksen, Stine Sofia Korreman^*

^*Corresponding author af dette arbejde

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avis › Tidsskriftartikel › Forskning › peer review

Abstract

Background and purpose: Interactive segmentation seeks to incorporate human knowledge into segmentation models and thereby reducing the total amount of editing of auto-segmentations. By performing only interactions which provide new information, segmentation performance may increase cost-effectively. The aim of this study was to develop, evaluate and test feasibility of a deep learning-based single-cycle interactive segmentation model with the input being computer tomography (CT) and a small amount of information rich contours. Methods and Materials: A single-cycle interactive segmentation model, which took CT and the most cranial and caudal contour slices for each of 16 organs-at-risk for head-and-neck cancer as input, was developed. A CT-only model served as control. The models were evaluated with Dice similarity coefficient, Hausdorff Distance 95th percentile and average symmetric surface distance. A subset of 8 organs-at-risk were selected for a feasibility test. In this, a designated radiation oncologist used both single-cycle interactive segmentation and atlas-based auto-contouring for three cases. Contouring time and added path length were recorded. Results: The medians of Dice coefficients increased with single-cycle interactive segmentation in the range of 0.004 (Brain)–0.90 (EyeBack_merged) when compared to CT-only. In the feasibility test, contouring time and added path length were reduced for all three cases as compared to editing atlas-based auto-segmentations. Conclusion: Single-cycle interactive segmentation improved segmentation metrics when compared to the CT-only model and was clinically feasible from a technical and usability point of view. The study suggests that it may be cost-effective to add a small amount of contouring input to deep learning-based segmentation models.

Originalsprog	Engelsk
Artikelnummer	100426
Tidsskrift	Physics and Imaging in Radiation Oncology
Vol/bind	26
Antal sider	6
DOI	https://doi.org/10.1016/j.phro.2023.100426
Status	Udgivet - apr. 2023

Adgang til dokumentet

10.1016/j.phro.2023.100426Licens: CC BY-NC-ND

Andre filer og links

Link to publication in Scopus

Citationsformater

@article{322f302173a0417f910abac6f71a90ea,

title = "A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer",

abstract = "Background and purpose: Interactive segmentation seeks to incorporate human knowledge into segmentation models and thereby reducing the total amount of editing of auto-segmentations. By performing only interactions which provide new information, segmentation performance may increase cost-effectively. The aim of this study was to develop, evaluate and test feasibility of a deep learning-based single-cycle interactive segmentation model with the input being computer tomography (CT) and a small amount of information rich contours. Methods and Materials: A single-cycle interactive segmentation model, which took CT and the most cranial and caudal contour slices for each of 16 organs-at-risk for head-and-neck cancer as input, was developed. A CT-only model served as control. The models were evaluated with Dice similarity coefficient, Hausdorff Distance 95th percentile and average symmetric surface distance. A subset of 8 organs-at-risk were selected for a feasibility test. In this, a designated radiation oncologist used both single-cycle interactive segmentation and atlas-based auto-contouring for three cases. Contouring time and added path length were recorded. Results: The medians of Dice coefficients increased with single-cycle interactive segmentation in the range of 0.004 (Brain)–0.90 (EyeBack_merged) when compared to CT-only. In the feasibility test, contouring time and added path length were reduced for all three cases as compared to editing atlas-based auto-segmentations. Conclusion: Single-cycle interactive segmentation improved segmentation metrics when compared to the CT-only model and was clinically feasible from a technical and usability point of view. The study suggests that it may be cost-effective to add a small amount of contouring input to deep learning-based segmentation models.",

keywords = "Auto-contouring, Auto-segmentation, Cancer, Deep learning, Head-and-neck, Interactive segmentation, Organs-at-risk, Radiotherapy, Single-cycle, Treatment planning",

author = "Rasmussen, {Mathis Ersted} and Nijkamp, {Jasper Albertus} and Eriksen, {Jesper Grau} and Korreman, {Stine Sofia}",

year = "2023",

month = apr,

doi = "10.1016/j.phro.2023.100426",

language = "English",

volume = "26",

journal = "Physics and Imaging in Radiation Oncology",

issn = "2405-6316",

publisher = "Elsevier",

}

A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer. / Rasmussen, Mathis Ersted ; Nijkamp, Jasper Albertus ; Eriksen, Jesper Grau et al.
I: Physics and Imaging in Radiation Oncology, Bind 26, 100426, 04.2023.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avis › Tidsskriftartikel › Forskning › peer review

TY - JOUR

T1 - A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer

AU - Rasmussen, Mathis Ersted

AU - Nijkamp, Jasper Albertus

AU - Eriksen, Jesper Grau

AU - Korreman, Stine Sofia

PY - 2023/4

Y1 - 2023/4

N2 - Background and purpose: Interactive segmentation seeks to incorporate human knowledge into segmentation models and thereby reducing the total amount of editing of auto-segmentations. By performing only interactions which provide new information, segmentation performance may increase cost-effectively. The aim of this study was to develop, evaluate and test feasibility of a deep learning-based single-cycle interactive segmentation model with the input being computer tomography (CT) and a small amount of information rich contours. Methods and Materials: A single-cycle interactive segmentation model, which took CT and the most cranial and caudal contour slices for each of 16 organs-at-risk for head-and-neck cancer as input, was developed. A CT-only model served as control. The models were evaluated with Dice similarity coefficient, Hausdorff Distance 95th percentile and average symmetric surface distance. A subset of 8 organs-at-risk were selected for a feasibility test. In this, a designated radiation oncologist used both single-cycle interactive segmentation and atlas-based auto-contouring for three cases. Contouring time and added path length were recorded. Results: The medians of Dice coefficients increased with single-cycle interactive segmentation in the range of 0.004 (Brain)–0.90 (EyeBack_merged) when compared to CT-only. In the feasibility test, contouring time and added path length were reduced for all three cases as compared to editing atlas-based auto-segmentations. Conclusion: Single-cycle interactive segmentation improved segmentation metrics when compared to the CT-only model and was clinically feasible from a technical and usability point of view. The study suggests that it may be cost-effective to add a small amount of contouring input to deep learning-based segmentation models.

AB - Background and purpose: Interactive segmentation seeks to incorporate human knowledge into segmentation models and thereby reducing the total amount of editing of auto-segmentations. By performing only interactions which provide new information, segmentation performance may increase cost-effectively. The aim of this study was to develop, evaluate and test feasibility of a deep learning-based single-cycle interactive segmentation model with the input being computer tomography (CT) and a small amount of information rich contours. Methods and Materials: A single-cycle interactive segmentation model, which took CT and the most cranial and caudal contour slices for each of 16 organs-at-risk for head-and-neck cancer as input, was developed. A CT-only model served as control. The models were evaluated with Dice similarity coefficient, Hausdorff Distance 95th percentile and average symmetric surface distance. A subset of 8 organs-at-risk were selected for a feasibility test. In this, a designated radiation oncologist used both single-cycle interactive segmentation and atlas-based auto-contouring for three cases. Contouring time and added path length were recorded. Results: The medians of Dice coefficients increased with single-cycle interactive segmentation in the range of 0.004 (Brain)–0.90 (EyeBack_merged) when compared to CT-only. In the feasibility test, contouring time and added path length were reduced for all three cases as compared to editing atlas-based auto-segmentations. Conclusion: Single-cycle interactive segmentation improved segmentation metrics when compared to the CT-only model and was clinically feasible from a technical and usability point of view. The study suggests that it may be cost-effective to add a small amount of contouring input to deep learning-based segmentation models.

KW - Auto-contouring

KW - Auto-segmentation

KW - Cancer

KW - Deep learning

KW - Head-and-neck

KW - Interactive segmentation

KW - Organs-at-risk

KW - Radiotherapy

KW - Single-cycle

KW - Treatment planning

UR - http://www.scopus.com/inward/record.url?scp=85151252819&partnerID=8YFLogxK

U2 - 10.1016/j.phro.2023.100426

DO - 10.1016/j.phro.2023.100426

M3 - Journal article

C2 - 37063613

AN - SCOPUS:85151252819

SN - 2405-6316

VL - 26

JO - Physics and Imaging in Radiation Oncology

JF - Physics and Imaging in Radiation Oncology

M1 - 100426

ER -

A simple single-cycle interactive strategy to improve deep learning-based segmentation of organs-at-risk in head-and-neck cancer

Abstract

Adgang til dokumentet

Andre filer og links

Fingeraftryk

Citationsformater