External elastic lamina vs. luminal diameter measurement for determining stent diameter by optical coherence tomography: an ILUMIEN III substudy

Aims Optical coherence tomography (OCT)-guided external elastic lamina (EEL)-based stent sizing is safe and as effective as intravascular ultrasound in achieving post-procedural lumen dimensions. However, when compared with automated lumen diameter (LD) measurements, this approach is time-consuming. We aimed to compare vessel diameter measurements and stent diameter selection using either of these approaches and examined whether applying a correction factor to automated LD measurements could result in selecting similar stent diameters to the EEL-based approach. Methods and results We retrospectively compared EEL-based measurements vs. automated LD in reference segments in 154 OCT acquisitions and derived a correction factor for stent sizing using the ratio of EEL to LD measurements. We then prospectively applied the correction factor in 119 OCT acquisitions. EEL could be adequately identified in 100 acquisitions (84%) at the distal reference to allow vessel diameter measurement. Vessel diameters were larger with EEL-based vs. LD measurements at both proximal (4.12 ± 0.74 vs. 3.14 ± 0.67 mm, P < 0.0001) and distal reference segments (3.34 ± 0.75 vs. 2.64 ± 0.65 mm, P < 0.0001). EEL-based downsizing led to selection of larger stents vs. an LD-based upsizing approach (3.33 ± 0.47 vs. 2.70 ± 0.44, P < 0.0001). Application of correction factors to LD [proximal 1.32 (IQR 1.23–1.37) and distal 1.25 (IQR 1.19–1.36)] resulted in discordance in stent sizing by >0.25 mm in 63% and potentially hazardous stent oversizing in 41% of cases. Conclusion EEL-based stent downsizing led to selection of larger stent diameters vs. LD upsizing. While applying a correction factor to automated LD measurements resulted in similar mean diameters to EEL-based measurements, this approach cannot be used clinically due to frequent and potentially hazardous stent over-sizing.

Comparison of retinal vessel diameter measurements from swept-source OCT angiography and adaptive optics ophthalmoscope

Background/imsTo compare the retinal vessel diameter measurements obtained from the swept-source optical coherence tomography angiography (OCTA; Plex Elite 9000, Carl Zeiss Meditec, USA) and adaptive optics ophthalmoscope (AOO; RTX1, Imagine Eyes, France).MethodsFifteen healthy subjects, 67% women, mean age (SD) 30.87 (6.19) years, were imaged using OCTA and AOO by a single experienced operator on the same day. Each eye was scanned using two OCTA protocols (3×3 mm2 and 9×9 mm2) and two to five AOO scans (1.2×1.2 mm2). The OCTA and AOO scans were scaled to the same pixel resolution. Two independent graders measured the vessel diameter at the same location on the region-of-interest in the three coregistered scans. Differences in vessel diameter measurements between the scans were assessed.ResultsThe inter-rater agreement was excellent for vessel diameter measurement in both OCTA protocols (ICC=0.92) and AOO (ICC=0.98). The measured vessel diameter was widest from the OCTA 3×3 mm2 (55.2±16.3 µm), followed by OCTA 9×9 mm2 (54.7±14.3 µm) and narrowest by the AOO (50.5±15.6 µm; p<0.001). Measurements obtained from both OCTA protocols were significantly wider than the AOO scan (OCTA 3×3 mm2: mean difference Δ=4.7 µm, p<0.001; OCTA 9×9 mm2: Δ=4.2 µm, p<0.001). For vessels >45 µm, it appeared to be larger in OCTA 3×3 mm2 scan than the 9×9 mm2 scan (Δ=1.9 µm; p=0.005), while vessels <45 µm appeared smaller in OCTA 3×3 mm2 scan (Δ=−1.3 µm; p=0.009)ConclusionsThe diameter of retinal vessels measured from OCTA scans were generally wider than that obtained from AOO scans. Different OCTA scan protocols may affect the vessel diameter measurements. This needs to be considered when OCTA measures such as vessel density are calculated.

Retinal Vessel Diameter Measurement Using Unsupervised Linear Discriminant Analysis

An automatic vessel diameter measurement technique based on linear discriminant analysis (LDA) has been proposed. After estimating the vessel wall, the vessel cross-section profile is divided into three regions: two corresponding to the background and one to the vessel. The algorithm was tested on more than 5000 cross-sections of retinal vessels from the REVIEW dataset through comparative study with the state-of-the-art techniques. Cross-correlation analyses were performed to determine the degree to which the proposed technique was close to the ground truth. The results indicate that proposed algorithm consistently performed better than most of other techniques and was highly correlated with the manual measurement as the reference diameter. The proposed method does not require any supervision and is suitable for automatic analysis.