Automatic analysis of bioresorbable vascular scaffolds in intravascular optical coherence tomography images

Abstract

The bioresorbable vascular scaffold (BVS) is a new generation of bioresorbable scaffold (BRS) for the treatment of coronary artery disease. A potential challenge of BVS is malapposition, which may possibly lead to late stent thrombosis. It is therefore important to conduct malapposition analysis right after stenting. Since an intravascular optical coherence tomography (IVOCT) image sequence contains thousands of BVS struts, manual analysis is labor intensive and time consuming. Computer-based automatic analysis is an alternative, but faces some difficulties due to the interference of blood artifacts and the uncertainty of the struts number, position and size. In this paper, we propose a novel framework for a struts malapposition analysis that breaks down the problem into two steps. Firstly, struts are detected by a cascade classifier trained by AdaBoost and a region of interest (ROI) is determined for each strut to completely contain it. Then, strut boundaries are segmented within ROIs through dynamic programming. Based on the segmentation result, malapposition analysis is conducted automatically. Tested on 7 pullbacks labeled by an expert, our method correctly detected 91.5% of 5821 BVS struts with 12.1% false positives. The average segmentation Dice coefficient for correctly detected struts was 0.81. The time consumption for a pullback is 15 on average. We conclude that our method is accurate and efficient for BVS strut detection and segmentation, and enables automatic BVS malapposition analysis in IVOCT images.

Automatic segmentation of optical coherence tomography pullbacks of coronary arteries treated with bioresorbable vascular scaffolds: Application to hemodynamics modeling

“…When considering only the ‘good’ classified OCT pullbacks, the validation of the stent strut detection algorithm resulted in a precision of ~91% and ~86% and a sensitivity of ~78% and ~76%, against the first end second expert image readers, respectively. The precision was similar to that found by Amrute et al [ 9 ] (93%) and Cao et al [ 8 ] (87.9%) and lower than that by Wang et al [ 10 ] (99.2% ± 0.1%). The sensitivity was lower than that found by Amrute et al [ 9 ] (90%), Cao et al [ 8 ] (91.5%) and Wang et al [ 10 ] (96.6% ± 2.0%).…”

“…The precision was similar to that found by Amrute et al [ 9 ] (93%) and Cao et al [ 8 ] (87.9%) and lower than that by Wang et al [ 10 ] (99.2% ± 0.1%). The sensitivity was lower than that found by Amrute et al [ 9 ] (90%), Cao et al [ 8 ] (91.5%) and Wang et al [ 10 ] (96.6% ± 2.0%). This last evidence can be partially explained by the fact that our detection algorithm was applied to OCT images with lower quality, in particular 8-bit images acquired with the Lunawave OCT imaging system (Terumo Corp.), as compared to the 16-bit images acquired with the C7-XR OCT system (St. Jude Medical) of the other studies [ 8 – 10 ].…”

“…The sensitivity was lower than that found by Amrute et al [ 9 ] (90%), Cao et al [ 8 ] (91.5%) and Wang et al [ 10 ] (96.6% ± 2.0%). This last evidence can be partially explained by the fact that our detection algorithm was applied to OCT images with lower quality, in particular 8-bit images acquired with the Lunawave OCT imaging system (Terumo Corp.), as compared to the 16-bit images acquired with the C7-XR OCT system (St. Jude Medical) of the other studies [ 8 – 10 ]. It is also worth noting that the comparison between our two manual segmentations resulted in a sensitivity of ~92% a precision of ~96% with high standard deviation (15% and 9%, respectively).…”

“…Nevertheless, these algorithms are not effective for polymeric bioresorbable scaffolds due to the different optical properties compared to the conventional metallic stents [ 7 ]. Few studies have proposed a segmentation strategy for polymeric bioresorbable scaffolds so far [ 8 – 10 ]. The limited research activity on the detection of those type of stents in OCT images may be addressed to the recent introduction of this technology into the market and the negative long-term clinical outcomes of the Absorb Bioresorbable Vascular Scaffold (BVS) (Abbott Vascular, Abbott Park, IL, USA) [ 11 ].…”

“…The currently available segmentation algorithms of polymeric bioresorbable scaffolds are characterized by good sensitivity and precision [ 8 – 10 ]. Their validation was performed on high quality 16-bit images obtained with the C7-XR OCT system (St. Jude Medical, St. Paul, MN, USA) [ 8 – 10 ] and typically on less than 10 pullbacks [ 8 , 10 ]. Little or nothing is known about the performance of automatic segmentation algorithms on 8-bit images acquired with different scanners.…”