Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients

Vivian Van Asperen; Josefien Van Den Berg; Fleur Lycklama; Victoria Marting; Sandra Cornelissen; Wim H. Van Zwam; Jeanette Hofmeijer; Aad Van Der Lugt; Theo Van Walsum; Matthijs Van Der Sluijs; Ruisheng Su

doi:10.1117/12.2606412

Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients

Vivian Van Asperen, Josefien Van Den Berg, Fleur Lycklama, Victoria Marting, Sandra Cornelissen, Wim H. Van Zwam, Jeanette Hofmeijer, Aad Van Der Lugt, Theo Van Walsum, Matthijs Van Der Sluijs, Ruisheng Su

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

3 Citations (Scopus)

133 Downloads (Pure)

Abstract

To develop an objective system for perfusion assessment in digital subtraction angiography (DSA), artery-vein (A/V) classification is essential. In this study, an automated A/V classification system in 2D DSA images of stroke patients is proposed. After preprocessing through vessel segmentation with a Frangi fitler and Gaussian smoothing, a time-intensity curve (TIC) of each vessel pixel was extracted and relevant parameters were calculated. Different combinations of input parameters were systematically tested to come to the optimal set of input parameters. The parameters formed the input for k-means (KM) and fuzzy c-means (FCM) clustering. Both algorithms were tested for clustering into 2 to 7 clusters. Cluster labeling was performed based on the average time to peak (TTP) of a cluster. A reference standard consisted of manually annotated DSA images of the MR CLEAN registry. Outcome measures were accuracy, true artery rate (TAR) and true vein rate (TVR). The optimal value for k was found to be 2 for both KM and FCM clustering. The optimal parameter set was: variance, standard deviation, maximal slope, peak width, time to peak, arrival time, maximal intensity and area under the TIC. No significant difference was found between FCM and KM clustering and. Both FCM and KM clustering yielded an average accuracy of 76%, average TAR of 74% and average TVR of 80%.

Original language	English
Title of host publication	Medical Imaging 2022
Subtitle of host publication	Image-Guided Procedures, Robotic Interventions, and Modeling
Editors	Cristian A. Linte, Jeffrey H. Siewerdsen
Publisher	SPIE
Volume	12034
ISBN (Electronic)	9781510649446
ISBN (Print)	978-1-5106-4943-9
DOIs	https://doi.org/10.1117/12.2606412
Publication status	Published - 4 Apr 2022
Event	SPIE Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling - Virtual, Online Duration: 21 Mar 2022 → 27 Mar 2022

Publication series

Name	Progress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume	12034
ISSN (Print)	1605-7422

Conference

Conference	SPIE Medical Imaging 2022
City	Virtual, Online
Period	21/03/22 → 27/03/22

Keywords

Artery-vein classification
DSA
Stroke
Vessels
22/3 OA procedure

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1117/12.2606412

120341NFinal published version, 3.5 MBLicence: Taverne

Cite this

Van Asperen, V., Van Den Berg, J., Lycklama, F., Marting, V., Cornelissen, S., Van Zwam, W. H., Hofmeijer, J., Van Der Lugt, A., Van Walsum, T., Van Der Sluijs, M., & Su, R. (2022). Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients. In C. A. Linte, & J. H. Siewerdsen (Eds.), Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling (Vol. 12034). Article 120341N (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 12034). SPIE. https://doi.org/10.1117/12.2606412

Van Asperen, Vivian ; Van Den Berg, Josefien ; Lycklama, Fleur et al. / Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients. Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling. editor / Cristian A. Linte ; Jeffrey H. Siewerdsen. Vol. 12034 SPIE, 2022. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).

@inproceedings{1f344a257ec44c49a399ac2c53e479d8,

title = "Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients",

abstract = "To develop an objective system for perfusion assessment in digital subtraction angiography (DSA), artery-vein (A/V) classification is essential. In this study, an automated A/V classification system in 2D DSA images of stroke patients is proposed. After preprocessing through vessel segmentation with a Frangi fitler and Gaussian smoothing, a time-intensity curve (TIC) of each vessel pixel was extracted and relevant parameters were calculated. Different combinations of input parameters were systematically tested to come to the optimal set of input parameters. The parameters formed the input for k-means (KM) and fuzzy c-means (FCM) clustering. Both algorithms were tested for clustering into 2 to 7 clusters. Cluster labeling was performed based on the average time to peak (TTP) of a cluster. A reference standard consisted of manually annotated DSA images of the MR CLEAN registry. Outcome measures were accuracy, true artery rate (TAR) and true vein rate (TVR). The optimal value for k was found to be 2 for both KM and FCM clustering. The optimal parameter set was: variance, standard deviation, maximal slope, peak width, time to peak, arrival time, maximal intensity and area under the TIC. No significant difference was found between FCM and KM clustering and. Both FCM and KM clustering yielded an average accuracy of 76%, average TAR of 74% and average TVR of 80%. ",

keywords = "Artery-vein classification, DSA, Stroke, Vessels, 22/3 OA procedure",

author = "{Van Asperen}, Vivian and {Van Den Berg}, Josefien and Fleur Lycklama and Victoria Marting and Sandra Cornelissen and {Van Zwam}, {Wim H.} and Jeanette Hofmeijer and {Van Der Lugt}, Aad and {Van Walsum}, Theo and {Van Der Sluijs}, Matthijs and Ruisheng Su",

note = "Funding Information: We thank the MR CLEAN Registry investigators for sharing the DSA image data. The MR CLEAN Registry was funded and carried out by the Erasmus University Medical Centre, Amsterdam UMC location AMC, and Maastricht University Medical Centre. The study was additionally funded by the Applied Scientific Institute for Neuromodulation (Toegepast Wetenschappelijk Instituut voor Neuromodulatie). Publisher Copyright: {\textcopyright} 2022 SPIE.; SPIE Medical Imaging 2022 : Image-Guided Procedures, Robotic Interventions, and Modeling ; Conference date: 21-03-2022 Through 27-03-2022",

year = "2022",

month = apr,

day = "4",

doi = "10.1117/12.2606412",

language = "English",

isbn = "978-1-5106-4943-9",

volume = "12034",

series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

publisher = "SPIE",

editor = "Linte, {Cristian A.} and Siewerdsen, {Jeffrey H.}",

booktitle = "Medical Imaging 2022",

address = "United States",

}

Van Asperen, V, Van Den Berg, J, Lycklama, F, Marting, V, Cornelissen, S, Van Zwam, WH, Hofmeijer, J, Van Der Lugt, A, Van Walsum, T, Van Der Sluijs, M & Su, R 2022, Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients. in CA Linte & JH Siewerdsen (eds), Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling. vol. 12034, 120341N, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 12034, SPIE, SPIE Medical Imaging 2022, Virtual, Online, 21/03/22. https://doi.org/10.1117/12.2606412

Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients. / Van Asperen, Vivian; Van Den Berg, Josefien; Lycklama, Fleur et al.
Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling. ed. / Cristian A. Linte; Jeffrey H. Siewerdsen. Vol. 12034 SPIE, 2022. 120341N (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 12034).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients

AU - Van Asperen, Vivian

AU - Van Den Berg, Josefien

AU - Lycklama, Fleur

AU - Marting, Victoria

AU - Cornelissen, Sandra

AU - Van Zwam, Wim H.

AU - Hofmeijer, Jeanette

AU - Van Der Lugt, Aad

AU - Van Walsum, Theo

AU - Van Der Sluijs, Matthijs

AU - Su, Ruisheng

N1 - Funding Information: We thank the MR CLEAN Registry investigators for sharing the DSA image data. The MR CLEAN Registry was funded and carried out by the Erasmus University Medical Centre, Amsterdam UMC location AMC, and Maastricht University Medical Centre. The study was additionally funded by the Applied Scientific Institute for Neuromodulation (Toegepast Wetenschappelijk Instituut voor Neuromodulatie). Publisher Copyright: © 2022 SPIE.

PY - 2022/4/4

Y1 - 2022/4/4

N2 - To develop an objective system for perfusion assessment in digital subtraction angiography (DSA), artery-vein (A/V) classification is essential. In this study, an automated A/V classification system in 2D DSA images of stroke patients is proposed. After preprocessing through vessel segmentation with a Frangi fitler and Gaussian smoothing, a time-intensity curve (TIC) of each vessel pixel was extracted and relevant parameters were calculated. Different combinations of input parameters were systematically tested to come to the optimal set of input parameters. The parameters formed the input for k-means (KM) and fuzzy c-means (FCM) clustering. Both algorithms were tested for clustering into 2 to 7 clusters. Cluster labeling was performed based on the average time to peak (TTP) of a cluster. A reference standard consisted of manually annotated DSA images of the MR CLEAN registry. Outcome measures were accuracy, true artery rate (TAR) and true vein rate (TVR). The optimal value for k was found to be 2 for both KM and FCM clustering. The optimal parameter set was: variance, standard deviation, maximal slope, peak width, time to peak, arrival time, maximal intensity and area under the TIC. No significant difference was found between FCM and KM clustering and. Both FCM and KM clustering yielded an average accuracy of 76%, average TAR of 74% and average TVR of 80%.

AB - To develop an objective system for perfusion assessment in digital subtraction angiography (DSA), artery-vein (A/V) classification is essential. In this study, an automated A/V classification system in 2D DSA images of stroke patients is proposed. After preprocessing through vessel segmentation with a Frangi fitler and Gaussian smoothing, a time-intensity curve (TIC) of each vessel pixel was extracted and relevant parameters were calculated. Different combinations of input parameters were systematically tested to come to the optimal set of input parameters. The parameters formed the input for k-means (KM) and fuzzy c-means (FCM) clustering. Both algorithms were tested for clustering into 2 to 7 clusters. Cluster labeling was performed based on the average time to peak (TTP) of a cluster. A reference standard consisted of manually annotated DSA images of the MR CLEAN registry. Outcome measures were accuracy, true artery rate (TAR) and true vein rate (TVR). The optimal value for k was found to be 2 for both KM and FCM clustering. The optimal parameter set was: variance, standard deviation, maximal slope, peak width, time to peak, arrival time, maximal intensity and area under the TIC. No significant difference was found between FCM and KM clustering and. Both FCM and KM clustering yielded an average accuracy of 76%, average TAR of 74% and average TVR of 80%.

KW - Artery-vein classification

KW - DSA

KW - Stroke

KW - Vessels

KW - 22/3 OA procedure

UR - http://www.scopus.com/inward/record.url?scp=85131955211&partnerID=8YFLogxK

U2 - 10.1117/12.2606412

DO - 10.1117/12.2606412

M3 - Conference contribution

AN - SCOPUS:85131955211

SN - 978-1-5106-4943-9

VL - 12034

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2022

A2 - Linte, Cristian A.

A2 - Siewerdsen, Jeffrey H.

PB - SPIE

T2 - SPIE Medical Imaging 2022

Y2 - 21 March 2022 through 27 March 2022

ER -

Van Asperen V, Van Den Berg J, Lycklama F, Marting V, Cornelissen S, Van Zwam WH et al. Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients. In Linte CA, Siewerdsen JH, editors, Medical Imaging 2022: Image-Guided Procedures, Robotic Interventions, and Modeling. Vol. 12034. SPIE. 2022. 120341N. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). doi: 10.1117/12.2606412

Automatic Artery/Vein Classification in 2D-DSA Images of Stroke Patients

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this