Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Standard

Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques. / Chu, Miao; Jia, Haibo; Gutiérrez-Chico, Juan Luis; Maehara, Akiko; Ali, Ziad A; Zeng, Xiaoling; He, Luping; Zhao, Chen; Matsumura, Mitsuaki; Wu, Peng; Zeng, Ming; Kubo, Takashi; Xu, Bo; Chen, Lianglong; Yu, Bo; Mintz, Gary S; Wijns, William; Holm, Niels Ramsing; Tu, Shengxian.

I: EuroIntervention, Bind 17, Nr. 1, 05.2021, s. 41-50.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Harvard

Chu, M, Jia, H, Gutiérrez-Chico, JL, Maehara, A, Ali, ZA, Zeng, X, He, L, Zhao, C, Matsumura, M, Wu, P, Zeng, M, Kubo, T, Xu, B, Chen, L, Yu, B, Mintz, GS, Wijns, W, Holm, NR & Tu, S 2021, 'Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques', EuroIntervention, bind 17, nr. 1, s. 41-50. https://doi.org/10.4244/EIJ-D-20-01355

APA

Chu, M., Jia, H., Gutiérrez-Chico, J. L., Maehara, A., Ali, Z. A., Zeng, X., He, L., Zhao, C., Matsumura, M., Wu, P., Zeng, M., Kubo, T., Xu, B., Chen, L., Yu, B., Mintz, G. S., Wijns, W., Holm, N. R., & Tu, S. (2021). Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques. EuroIntervention, 17(1), 41-50. https://doi.org/10.4244/EIJ-D-20-01355

CBE

Chu M, Jia H, Gutiérrez-Chico JL, Maehara A, Ali ZA, Zeng X, He L, Zhao C, Matsumura M, Wu P, Zeng M, Kubo T, Xu B, Chen L, Yu B, Mintz GS, Wijns W, Holm NR, Tu S. 2021. Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques. EuroIntervention. 17(1):41-50. https://doi.org/10.4244/EIJ-D-20-01355

MLA

Vancouver

Chu M, Jia H, Gutiérrez-Chico JL, Maehara A, Ali ZA, Zeng X o.a. Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques. EuroIntervention. 2021 maj;17(1):41-50. https://doi.org/10.4244/EIJ-D-20-01355

Author

Chu, Miao ; Jia, Haibo ; Gutiérrez-Chico, Juan Luis ; Maehara, Akiko ; Ali, Ziad A ; Zeng, Xiaoling ; He, Luping ; Zhao, Chen ; Matsumura, Mitsuaki ; Wu, Peng ; Zeng, Ming ; Kubo, Takashi ; Xu, Bo ; Chen, Lianglong ; Yu, Bo ; Mintz, Gary S ; Wijns, William ; Holm, Niels Ramsing ; Tu, Shengxian. / Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques. I: EuroIntervention. 2021 ; Bind 17, Nr. 1. s. 41-50.

Bibtex

@article{e8b32c50345f47cc80d552c2b3746e05,
title = "Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques",
abstract = "BACKGROUND: Intravascular optical coherence tomography (IVOCT) enables detailed plaque characterisation in-vivo, but visual assessment is time-consuming and subjective.AIMS: This study aims to develop and validate an automatic framework for IVOCT plaque characterisation using artificial intelligence (AI).METHODS: IVOCT pullbacks from 5 international centres were analysed in a corelab, annotating basic plaque components, inflammatory markers and other structures. A deep convolutional network with encoding-decoding architecture and pseudo-3D input was developed and trained using hybrid loss. The proposed network was integrated into commercial software to be externally validated on additional IVOCT pullbacks from three international corelabs, taking the consensus among corelabs as reference.RESULTS: Annotated images from 509 pullbacks (391 patients) were divided into 10,517 and 1,156 cross-sections for the training and testing datasets, respectively. Dice coefficient of the model was 0.906 for fibrous plaque, 0.848 for calcium and 0.772 for lipid in the testing dataset. Excellent agreement in plaque burden quantification was observed between the model and manual measurements (R2=0.98). In the external validation, the software correctly identified 518 out of 598 plaque regions from 300 IVOCT cross-sections, with a diagnostic accuracy of 97.6%[95%CI:93.4%-99.3%] in fibrous plaque, 90.5%[95%CI:85.2%-94.1%] in lipid and 88.5%[95%CI:82.4%-92.7%] in calcium. The median time required for analysis was 21.4 (18.6-25.0) seconds per pullback.CONCLUSIONS: A novel AI framework for automatic plaque characterisation in IVOCT was developed, providing excellent diagnostic accuracy in both internal and external validation. This model might reduce subjectivity in image interpretation and facilitate IVOCT quantification of plaque composition, with potential applications in research and IVOCT-guided PCI.",
author = "Miao Chu and Haibo Jia and Guti{\'e}rrez-Chico, {Juan Luis} and Akiko Maehara and Ali, {Ziad A} and Xiaoling Zeng and Luping He and Chen Zhao and Mitsuaki Matsumura and Peng Wu and Ming Zeng and Takashi Kubo and Bo Xu and Lianglong Chen and Bo Yu and Mintz, {Gary S} and William Wijns and Holm, {Niels Ramsing} and Shengxian Tu",
year = "2021",
month = may,
doi = "10.4244/EIJ-D-20-01355",
language = "English",
volume = "17",
pages = "41--50",
journal = "EuroIntervention",
issn = "1774-024X",
publisher = "Europa Digital & Publishing",
number = "1",

}

RIS

TY - JOUR

T1 - Artificial intelligence and optical coherence tomography for the automatic characterisation of human atherosclerotic plaques

AU - Chu, Miao

AU - Jia, Haibo

AU - Gutiérrez-Chico, Juan Luis

AU - Maehara, Akiko

AU - Ali, Ziad A

AU - Zeng, Xiaoling

AU - He, Luping

AU - Zhao, Chen

AU - Matsumura, Mitsuaki

AU - Wu, Peng

AU - Zeng, Ming

AU - Kubo, Takashi

AU - Xu, Bo

AU - Chen, Lianglong

AU - Yu, Bo

AU - Mintz, Gary S

AU - Wijns, William

AU - Holm, Niels Ramsing

AU - Tu, Shengxian

PY - 2021/5

Y1 - 2021/5

N2 - BACKGROUND: Intravascular optical coherence tomography (IVOCT) enables detailed plaque characterisation in-vivo, but visual assessment is time-consuming and subjective.AIMS: This study aims to develop and validate an automatic framework for IVOCT plaque characterisation using artificial intelligence (AI).METHODS: IVOCT pullbacks from 5 international centres were analysed in a corelab, annotating basic plaque components, inflammatory markers and other structures. A deep convolutional network with encoding-decoding architecture and pseudo-3D input was developed and trained using hybrid loss. The proposed network was integrated into commercial software to be externally validated on additional IVOCT pullbacks from three international corelabs, taking the consensus among corelabs as reference.RESULTS: Annotated images from 509 pullbacks (391 patients) were divided into 10,517 and 1,156 cross-sections for the training and testing datasets, respectively. Dice coefficient of the model was 0.906 for fibrous plaque, 0.848 for calcium and 0.772 for lipid in the testing dataset. Excellent agreement in plaque burden quantification was observed between the model and manual measurements (R2=0.98). In the external validation, the software correctly identified 518 out of 598 plaque regions from 300 IVOCT cross-sections, with a diagnostic accuracy of 97.6%[95%CI:93.4%-99.3%] in fibrous plaque, 90.5%[95%CI:85.2%-94.1%] in lipid and 88.5%[95%CI:82.4%-92.7%] in calcium. The median time required for analysis was 21.4 (18.6-25.0) seconds per pullback.CONCLUSIONS: A novel AI framework for automatic plaque characterisation in IVOCT was developed, providing excellent diagnostic accuracy in both internal and external validation. This model might reduce subjectivity in image interpretation and facilitate IVOCT quantification of plaque composition, with potential applications in research and IVOCT-guided PCI.

AB - BACKGROUND: Intravascular optical coherence tomography (IVOCT) enables detailed plaque characterisation in-vivo, but visual assessment is time-consuming and subjective.AIMS: This study aims to develop and validate an automatic framework for IVOCT plaque characterisation using artificial intelligence (AI).METHODS: IVOCT pullbacks from 5 international centres were analysed in a corelab, annotating basic plaque components, inflammatory markers and other structures. A deep convolutional network with encoding-decoding architecture and pseudo-3D input was developed and trained using hybrid loss. The proposed network was integrated into commercial software to be externally validated on additional IVOCT pullbacks from three international corelabs, taking the consensus among corelabs as reference.RESULTS: Annotated images from 509 pullbacks (391 patients) were divided into 10,517 and 1,156 cross-sections for the training and testing datasets, respectively. Dice coefficient of the model was 0.906 for fibrous plaque, 0.848 for calcium and 0.772 for lipid in the testing dataset. Excellent agreement in plaque burden quantification was observed between the model and manual measurements (R2=0.98). In the external validation, the software correctly identified 518 out of 598 plaque regions from 300 IVOCT cross-sections, with a diagnostic accuracy of 97.6%[95%CI:93.4%-99.3%] in fibrous plaque, 90.5%[95%CI:85.2%-94.1%] in lipid and 88.5%[95%CI:82.4%-92.7%] in calcium. The median time required for analysis was 21.4 (18.6-25.0) seconds per pullback.CONCLUSIONS: A novel AI framework for automatic plaque characterisation in IVOCT was developed, providing excellent diagnostic accuracy in both internal and external validation. This model might reduce subjectivity in image interpretation and facilitate IVOCT quantification of plaque composition, with potential applications in research and IVOCT-guided PCI.

U2 - 10.4244/EIJ-D-20-01355

DO - 10.4244/EIJ-D-20-01355

M3 - Journal article

C2 - 33528359

VL - 17

SP - 41

EP - 50

JO - EuroIntervention

JF - EuroIntervention

SN - 1774-024X

IS - 1

ER -