EUVIMEDEUVIMED entdecken
Vergleichsstudie

Deep learning-based plaque characterization in hybrid IVUS-OCT images is superior to single-modality deep learning analysis and human experts: head-to-head comparison against histology

Retesh Bajaj iD, Xingru Huang, Natasha Alves-Kotzev, Jill J Weyers, Molly Levine, Mohil Garg iD, Mohamed Mohamed, Soe Maung, Ramya Parasa, Murat Çap, Ryo Torii iD, Rob Krams iD und 9 weitere

Cardiovascular Research · 16.03.2026 · 122 (2) · 245-257

DOI: 10.1093/cvr/cvaf281 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Cardiovascular Research, online erschienen am 28.01.2026. Verfasst von Retesh Bajaj und 20 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cardiology, St Michael's Hospital, Toronto, Canada. Hauptthemen laut MEDLINE: Plaque, Atherosclerotic, Tomography, Optical Coherence, Ultrasonography, Interventional, Deep Learning. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 73 Literaturangaben. Gefördert von: British Heart Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

Hybrid intravascular ultrasound-optical coherence tomography (IVUS-OCT) can enable more accurate plaque characterization than single-modality intravascular imaging, enhancing treatment planning and vulnerable plaque detection. However, image interpretation in IVUS-OCT is challenging and time-consuming. To overcome this limitation, we introduce a novel histology-trained deep learning (DL)-classifier for plaque component classification in IVUS-OCT images and compare its performance against single-modality DL and expert analysts. Methods and results IVUS-OCT frames and matched histological sections from 10 cadaveric human hearts were included in this analysis. The histological data were used to define fibrotic, calcific, and necrotic core tissue regions of interest (ROIs) in IVUS-OCT and used to train three DL-classifiers for IVUS, OCT, or hybrid IVUS-OCT image analysis (992 frames) and test their performance (264 frames). The test set was additionally annotated by experts from three different core labs, and their estimations and those of the DL-classifiers were compared with histology. The IVUS-OCT DL-classifier had a superior performance to the IVUS-DL, OCT-DL, and the expert analysts in detecting plaque phenotypes (Kappa 0.60 vs. 0.19, 0.35, and 0.53, respectively) and accurately classified 68% of histologically defined fibroatheromas. The hybrid IVUS-OCT DL-classifier also had a better performance than single-modality DL-classifiers and the experts in assessing tissue types in ROIs annotated by histology (overall accuracy 86.7% compared with 73.2% for IVUS-DL, 66.6% for OCT-DL, and 70.6% for the experts).

Conclusion

Plaque characterization using a histology-trained hybrid IVUS-OCT DL-classifier is feasible and enables more accurate detection of plaque components and phenotype classification than single-modality DL-classifiers and expert analysts.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Plaque, AtheroscleroticTomography, Optical CoherenceUltrasonography, InterventionalDeep LearningCoronary VesselsImage Interpretation, Computer-AssistedCoronary Artery DiseaseMultimodal ImagingVascular CalcificationHumansPredictive Value of TestsFibrosisReproducibility of ResultsNecrosisCadaverObserver VariationClassification Algorithms

Stichwörter der Autor:innen

Intravascular ultrasoundMachine learningOptical coherence tomographyPlaque characterization

Autor:innen und Institutionen

  • Department of Cardiology, St Michael's Hospital, Toronto, Canada.
  • Division of Cardiology, University of Toronto, Toronto, Canada.
  • Cardiovascular Devices Hub, Centre for Cardiovascular Medicine and Devices, William Harvey Research Institute, Queen Mary University of London, London, UK.
  • School of Electronic Engineering and Computer Science, Queen Mary University of London, London, UK.
  • Sunnybrook Health Sciences Centre, Toronto, Canada.
  • Department of Cardiology, Medstar Cardiovascular Research Network, Medstar Washington Hospital Center, Washington, DC, USA.

ORCID-Profile: Retesh Bajaj, Mohil Garg, Ryo Torii, Rob Krams, Flavio Giuseppe Biccirè, Anthony Mathur, Andreas Baumbach, Qianni Zhang

Förderung

  • British Heart Foundation

Zitieren

APA

Bajaj, R., Huang, X., Alves-Kotzev, N., Weyers, J. J., Levine, M., Garg, M., Mohamed, M., Maung, S., Parasa, R., Çap, M., Torii, R., Krams, R., Butany, J., Biccirè, F. G., Garcia-Garcia, H., Raber, L., Mathur, A., Baumbach, A., Zhang, Q., …, & Bourantas, C. V. (2026). Deep learning-based plaque characterization in hybrid IVUS-OCT images is superior to single-modality deep learning analysis and human experts: head-to-head comparison against histology. Cardiovascular Research, 122(2), 245-257. https://doi.org/10.1093/cvr/cvaf281

Vancouver

Bajaj R, Huang X, Alves-Kotzev N, Weyers JJ, Levine M, Garg M, et al. Deep learning-based plaque characterization in hybrid IVUS-OCT images is superior to single-modality deep learning analysis and human experts: head-to-head comparison against histology. Cardiovascular Research. 2026;122(2):245-257. doi:10.1093/cvr/cvaf281

BibTeX
@article{Bajaj2026,
  author = {Bajaj, Retesh and Huang, Xingru and Alves-Kotzev, Natasha and Weyers, Jill J and Levine, Molly and Garg, Mohil and Mohamed, Mohamed and Maung, Soe and Parasa, Ramya and Çap, Murat and Torii, Ryo and Krams, Rob and Butany, Jagdish and Biccirè, Flavio Giuseppe and Garcia-Garcia, Hector and Raber, Lorenz and Mathur, Anthony and Baumbach, Andreas and Zhang, Qianni and Courtney, Brian K and Bourantas, Christos V},
  title = {Deep learning-based plaque characterization in hybrid IVUS-OCT images is superior to single-modality deep learning analysis and human experts: head-to-head comparison against histology},
  journal = {Cardiovascular Research},
  year = {2026},
  volume = {122},
  number = {2},
  pages = {245--257},
  doi = {10.1093/cvr/cvaf281}
}
Kostenlos starten →