Author + information
- Vikram Baruah1,
- Aydin Zahedivash2,
- Hoyt Taylor3,
- Austin McElroy4,
- Deborah Vela5,
- L. Maximilian Buja6,
- Thomas Milner7 and
- Marc Feldman8
- 1University of Texas at Austin, Austin, Texas, United States
- 2Dell Medical School, University of Texas, Austin, Texas, United States
- 3University of Texas Health Science Center, San Antonio, Texas, United States
- 4University of Texas at Austin
- 5Texas Heart Institute, Houston, Texas, United States
- 6The University of Texas HSC
- 7Austin, Texas, United States
- 8University of Texas Health Science Center at San Antonio, San Antonio, Texas, United States
Although virtual histology algorithms using intravascular ultrasound data can successfully classify plaque composition, parallel results have not been demonstrated with intravascular optical coherence tomography (IVOCT). The complexity of IVOCT images has limited its clinical acceptance despite its greater resolution. For example, thin-capped fibroatheromas (TCFAs), which are uniquely recognized by IVOCT, can be falsely identified by expert viewers.
Neural network features and nodes were optimized to best classify arterial fibrous, calcium and lipid tissue in IVOCT images. Lipid pixels were used to isolate TCFA. The network was trained on pixels sampled from 21 plaque lesions in 11 left anterior descending and 4 right coronary arteries. Imaging was conducted on 11 human hearts (3 women, 8 men) within 24 hours of death. Age at death was 65 ± 11 years. Accuracy was calculated by comparing network classification to histological assessment.
Fibrous, calcium, lipid, and TCFA pixels were classified with 92.4%, 89.7%, 95% accuracy respectively (Figure 1). Accuracies reported herein exceed previously reported values for automated IVOCT plaque classification.
We have developed a histology-validated IVOCT-based automated plaque classification algorithm that is able to colorize plaque composition in human coronary arteries. Gaining spatial insight into the composition of plaque can enable precise TCFA identification and provides a valuable diagnostic methodology to assess the efficacy of therapeutic interventions.