Time-domain Optical Coherence Tomography Images Research Articles

OCT Signal Enhancement with Deep Learning

To establish whether deep learning methods are able to improve the signal-to-noise ratio of time-domain (TD) OCT images to approach that of spectral-domain (SD) OCT images. Method agreement study and progression detection in a randomized, double-masked, placebo-controlled, multicenter trial for open-angle glaucoma (OAG), the United Kingdom Glaucoma Treatment Study (UKGTS). The training and validation cohort comprised 77 stable OAG participants with TD OCT and SD OCT imaging at up to 11 visits within 3 months. The testing cohort comprised 284 newly diagnosed OAG patients with TD OCT images from a cohort of 516 recruited at 10 United Kingdom centers between 2007 and2010. An ensemble of generative adversarial networks (GANs) was trained on TD OCT and SD OCT image pairs from the training dataset and applied to TD OCT images from the testing dataset. Time-domain OCT images were converted to synthesized SD OCT images and segmented via Bayesian fusion on the output of the GANs. Bland-Altman analysis assessed agreement between TD OCT and synthesized SD OCT average retinal nerve fiber layer thickness (RNFLT) measurements and the SD OCT RNFLT. Analysis of the distribution of the rates of RNFLT change in TD OCT and synthesized SD OCT in the two treatment arms of the UKGTS was compared. A Cox model for predictors of time-to-incident visual field (VF) progression was computed with the TD OCT and the synthesized SD OCT images. The 95% limits of agreement were between TD OCT and SD OCT were 26.64 to -22.95; between synthesized SD OCT and SD OCT were 8.11 to -6.73; and between SD OCT and SD OCT were 4.16 to -4.04. The mean difference in the rate of RNFLT change between UKGTS treatment and placebo arms with TD OCT was 0.24 (P= 0.11) and with synthesized SD OCT was 0.43 (P= 0.0017). The hazard ratio for RNFLT slope in Cox regression modeling for time to incident VF progression was 1.09 (95% confidence interval [CI], 1.02-1.21; P=0.035) for TD OCT and 1.24 (95% CI, 1.08-1.39; P= 0.011) for synthesized SD OCT. Image enhancement significantly improved the agreement of TD OCT RNFLT measurements with SD OCT RNFLT measurements. The difference, and its significance, in rates of RNFLT change in the UKGTS treatment arms was enhanced and RNFLT change became a stronger predictor of VF progression.

Imaging Modality in Diagnosis and Monitoring of Glaucoma: Spectral Domain Optical Coherence Tomography

To provide an overview of spectral domain optical coherence tomography (SDOCT) technology for clinical practice. Since glaucoma is the leading cause of irreversible blindness, early diagnosis and detection of progression are important. Early treatment can be sight saving. Currently, SDOCT is the most commonly used high-resolution imaging modality in diagnosing and monitoring glaucoma. SDOCT focuses on three anatomic regions: the retinal nerve fiber layer, the optic nerve head, and the macular ganglion cell region. Each anatomic region can be objectively quantified, and many of these measurements can be compared to a normative database and then classified into within normal limits, borderline, or outside normal limits. The advantages and disadvantages of the different scan regions are described, and future directions and needs in glaucoma imaging are discussed. SDOCT affords improvements over time domain OCT imaging for glaucoma care, but future research is needed.

Anatomy of the fetal membranes using optical coherence tomography: Part 1

IntroductionIn vitro studies on the structure of human fetal membranes have involved light or electron microscopy with fixation, dehydration, and staining. Recently, optical coherence tomography (OCT), an imaging technology, has provided high-resolution cross-sectional images of living biological tissues, with a penetration of 2–3 mm. We evaluated the use of this technology to examine the histologic features of human fetal membranes immediately after delivery. MethodsSamples of fetal membranes of ten patients undergoing cesarean deliveries (four uncomplicated pregnancies, four with preeclampsia, and two with chorioamnionitis) and eight patients undergoing vaginal deliveries (six uncomplicated pregnancies and two with chorioamnionitis) were collected immediately after delivery. Samples were stretched across customized disks, rinsed, and analyzed using a time-domain OCT imaging system. Following OCT scanning, the samples were placed in formalin for histologic study. The OCT images were compared to histologic images of common human fetal membrane features. ResultsWe were able to delineate the layers of the fetal membranes using bench-top time-domain OCT. The system was able to image histologic features of the fetal membranes, such as microscopic chorionic pseudocysts, ghost villi, meconium stained membranes, and chorioamnionitis. The OCT images corresponded with the histologic findings. DiscussionThis feasibility study demonstrates the potential of OCT technology for real-time assessment of human fetal membranes and may provide clinically useful information at delivery.

Predictive Value in Retinal Vein Occlusions of Early Versus Late or Incomplete Ranibizumab Response Defined by Optical Coherence Tomography

To determine if optical coherence tomography (OCT) at baseline or month 3 in the Treatment of Macular Edema following Branch Retinal Vein Occlusion: Evaluation of Efficacy and Safety (BRAVO) and Treatment of Macular Edema following Central Retinal Vein Occlusion: Evaluation of Efficacy and Safety (CRUISE) studies provides information that predicts visual outcome. Post hoc analysis from 2 prospective, randomized, controlled clinical trials. Three hundred ninety-seven patients from the BRAVO study and 392 patients from the CRUISE study. Time-domain OCT imaging data were analyzed. Mean change from baseline best-corrected visual acuity (BCVA) letter score at month 6 and month 12. Among ranibizumab-treated patients, 71.2% (0.3 mg) and 78.5% (0.5 mg) in the CRUISE study and 79.1% (0.3 mg) and 84.7% (0.5 mg) in the BRAVO study had central foveal thickness (CFT) of 250 μm or less at month 3 and therefore were categorized as early ranibizumab responders. Early ranibizumab responders had excellent visual outcomes regardless of ranibizumab dose; mean improvement in BCVA letter score at 6 and 12 months was 15.0 to 16.5 (central retinal vein occlusion [CRVO]) and 17.4 to 19.1 (branch retinal vein occlusion [BRVO]). Late or incomplete ranibizumab responders with CRVO (CFT >250 μm at month 3) did not fare as well as early responders if they were treated with 0.3 mg ranibizumab (month 6, P = 0.012). At month 6, compared with ranibizumab-treated CRVO patients with resolved cystoid macular edema (CME) at month 3, those with persistent CME did worse, on average, and significantly so for 0.5 mg (13.1 vs. 18.6; P = 0.027). At baseline, subretinal fluid (SRF) was present in 57% of patients with CRVO and in 45% of patients with BRVO; its presence did not portend a poor outcome in patients treated with ranibizumab for whom SRF was eliminated in almost all by month 3. At month 3 of ranibizumab treatment, OCT images provide predictive information for patients with CRVO, but not for those with BRVO. Visual outcome at months 6 and 12 was reduced in 0.5 mg ranibizumab-treated patients with CRVO who had persistent CME at month 3. It also was reduced in CRVO for those with CFT of more than 250 μm at month 3 who were treated with 0.3 mg ranibizumab. The findings suggest that late or incomplete responders may need careful follow-up. Proprietary or commercial disclosure may be found after the references.

Predominant microvessel proliferation in coronary stent restenotic tissue in patients with diabetes: Insights from optical coherence tomography image analysis

BackgroundCoronary optical coherence tomography (OCT) enables virtual depiction of histological findings of in-stent restenotic tissue. The aim of this study was to investigate the microvessel proliferation within in-stent restenotic tissue and the influence of diabetes mellitus (DM). MethodsWe examined 54 in-stent restenotic coronary artery lesions (stenotic area>50%) from 50 consecutive patients including 28 with DM (56%) and 9 insulin-treated DM patients (18%); who underwent coronary time-domain OCT imaging with automatic pull back (1mm/s, 20 frames/s). Microvessels were defined as low-signal cavities with a diameter of 50–150 microns and a trajectory parallel to the lumen recognized on 3 consecutive cross-sectional OCT image frames. The microvessel index was calculated as the number of frames with microvessel/total number of frames×100. Patients were stratified into 3 groups: 1) without microvessels, 2) with a low (< median value) microvessel index, 3) with a high microvessel index. ResultsMicrovessels were detected in 566 frames (3.1%) from 26 lesions (48%) in 24 patients (48%). A greater incidence of DM and higher serum glucose levels were observed in the high microvessel index group (DM: 42% vs 58% vs 83%, p=0.049; serum glucose level: 118.2±44.6 vs 122.6±31.0 vs 172.8±63.1mg/dL, p=0.03 between low and high microvessel index group, p=0.005 between no microvessel and high microvessel index group). ConclusionsMicrovessel formation may be a unique pathophysiological factor of in-stent restenoses in patients with DM.

Retinal measurements using time domain OCT imaging before and after myopic Lasik

Citation information: Lei F, Burns SA, Shao L &amp; Yang Y. Retinal measurements using time domain OCT imaging before and after myopic Lasik. Ophthalmic Physiol Opt 2012, 32, 222–227. doi: 10.1111/j.1475‐1313.2012.00905.xAbstractPurpose: To compare retinal measurements obtained by time domain optical coherence tomography (OCT) devices before and after myopic laser in situ keratomileusis (Lasik) and to assess the interaction of Lasik and retinal structures as measured by time domain OCT.Methods: Fifty‐three patients randomly selected participated in the study. Only the right eye of each subject was included in the study. Comprehensive ophthalmic examinations including refraction examination, slit lamp examination, dilated fundus examination, corneal topography, corneal thickness, intraocular pressure, and retinal Stratus OCT scans were acquired for each patient before myopic Lasik and 3 months after surgery.Results: Total macular volume (TMV) changed significantly between preoperative and postoperative measurements (p = 0.003). No statistical differences were found between preoperative and postoperative disc area, rim area, cup/disk vert. ratio, or average foveal thickness (p &gt; 0.05). The variation in TMV correlated significantly with the change in spherical refraction equivalent, maximal corneal curvature, minimal corneal curvature, and corneal ablation depth.Conclusions: Most retinal OCT measurements undergo no obvious changes after myopic Lasik. The increased TMV measurements we measured after Lasik seem to be correlated with the alteration in corneal shape. The exact mechanism for this change is not clear, while we examined several possibilities including subclinical macular oedema, magnification changes, errors in OCT analysis and IOP, none of these seem to be a likely cause.

Abstract 9330: In-stent Restenotic Tissue with a Microchannel in Patients with Diabetes: Insights from Coronary Optical Coherence Tomography Image Analysis

Introduction Coronary optical coherence tomography (OCT) allows the detection of subtle differences in patients with coronary artery disease. To date, OCT can virtually depict the various histological findings that occur at in-stent restenotic tissue. The present study investigated the clinical implications of microchannels located in in-stent restenotic sites. Methods We examined 48 consecutive in-stent restenotic (stenosis area, &gt;50%) coronary artery lesions from 44 patients who underwent coronary OCT. OCT images were obtained with a time-domain OCT imaging system. The images were acquired during automatic-pull back (1 mm/s; 20 frames/s). A microchannel was defined as the low signal cavity with a diameter of 50-150 µm (Figure; white arrows) and was evaluated by 2 independent analysts. A third analyst provided the final decision when there was a discrepancy between the first and second analysts. Diabetes was diagnosed in accordance with the guidelines of the Japanese Diabetes Society. Results In total, 16,008 frames were evaluated. Microchannels were detected in 464 (2.9%) frames from 24 (50%) lesions in 22 (50%) patients. The kappa statistic on the lesion level between the 2 analysts was 0.875. Diabetes was diagnosed in 23 (52.3%) patients, and 7 (15.9%) patients had insulin therapy. The incidence of microchannels was significantly more frequent in patients with diabetes (65.2% vs. 33.3%, p = 0.03). In contrast, none of the coronary risk factors (hypertension, dyslipidemia, smoking) other than diabetes showed a significant relationship with microchannel formation. Conclusion Microchannels were frequently observed at in-stent restenotic tissue sites in patients with diabetes. The results indicated that microvessel formation, which is a common diabetic complication, is a unique pathophysiological factor of in-stent restenosis in patients with diabetes. Neovascular formation may be targeted to prevent in-stent restenosis in patients with diabetes.

Coronary perforation during conventional time domain optical coherence tomography

Coronary artery perforation is a rare, but well recognized, compli-cation of percutaneous coronary intervention (PCI), with a reportedincidence of 0.14–0.35%, but with a mortality as high as 10% [1].Optical coherence tomography (OCT) has emerged as a powerfultool for use in stent assessment, and, within a short time, it hasbecome the modality of choice for study of stent and vascularinteractions in vivo. However, in a first generation pullback Time-Domain OCT imaging wire system (TD-OCT; M2 OCT systems andImageWire, LightLab Imaging, Inc, Westford, MA), blood clearancewas achievedby proximal balloonocclusion andflushingof the vesselusing the distal end holes of the occlusion balloon (Helios OcclusionBalloon Catheter, Goodman-Avantec Vascular Corp, Sunnyvale, CA).We report on a case of coronary perforation during first generationTD-OCT imaging as routine follow-up angiography.A 64-year-old woman with a history of previous percutaneouscoronary intervention (PCI) was admitted for follow-up coronaryangiography and OCT imaging. She had undergone PCI at the leftanterior descending artery (LAD) 1 year ago and had no anginalsymptoms after PCI. Cardiac examination was unremarkable, with aheart rate of 72 bpm and blood pressure (BP) 130/80 mm Hg.Diagnostic coronary angiogram, which was performed via the righttransradial approach using 6 Fr equipment, revealed no in-stentrestenosis of previously implanted stents with moderate distal leftanterior descending(LAD)artery stenosis (Fig. 1). TD-OCT imaging wasperformedusingaroutineprocedure,suchasproximalballoonocclusionatproximalLADand flushingwithRinger'slactateat1 ml/s.DuringOCT,abrupt BP depression and loss of consciousness developed. Followingimmediate removal of the OCT device and occluding balloon, coronaryangiography showed a huge coronary perforation with extravasation atthe distal LAD-D2 bifurcated lesion. We reinserted a guidewire and aconventionalballoonwasinflated just proximal to thelesion in order tosealtheperforation.Resuscitationwithinotropesandpericardiocentesiswere performed.The contrast leakcontinued inspiteofalongperiodofballoon inflation; therefore, a 3.0×15 mm Coronary Stent Graft™ wasinserted,butfailedtosealtheperforationduetoasmallcoronaryarteryand bulky graft stent (Fig. 2). The patient was transferred to theoperating room for surgical repair. Following sternotomy, a hugeperforated artery measuring approximately 6 –7mm was observed.Leftinternalmammaryartery(LIMA)graftstoLADandasaphenousveingrafttodiagonalwereplaced.Followingsuturingofthepericardiumandsternum, the patient returned to the CCU. She expired due to suddencardiac arrest at 7 days after surgery.OCT is high resolution, and, combined with the blood-free environ-ment, provides a substantial advantage for the study of coronary stent,when compared with previous intravascular ultrasound (IVUS) [2].Thistool enables detailed evaluation of tissue coverage and stent malapposi-tion in follow-up analysis. This specific parameter has been used as a

Influence of tooth-surface hydration conditions on optical coherence-tomography imaging

ObjectivesThis study examined the influence of tooth-surface hydration conditions on optical coherence tomography (OCT) imaging. MethodsEmitted light was coupled in a single-mode fibre-optic Michelson interferometer, and delivered to a sample and a reference mirror mounted on a linearly translating galvanometer driven by a triangular voltage waveform at a fringe-modulation frequency of 1kHz. Backscattered light from the sample was coupled back to the system, digitised and used to create two-dimensional images together with beam scanning. OCT imaging of the occlusal surfaces of 10 extracted human teeth was performed soon after mounting (‘wet’ condition), immediately after air blowing for 10s (‘air-blow’ condition), and at intervals thereafter (‘1-min’, ‘5-min’ and ‘10-min’ conditions, respectively). ResultsFor the wet condition, three distinct peaks indicated the water surface, tooth surface and dentino-enamel junction. Backscattered light was detected at levels above noise, possibly causing grainy OCT images. By contrast, two distinct peaks indicating the tooth surface and dentino-enamel junction were observed for the air-blow and 10-min conditions, with greater signal intensities for the latter. The intensity was lower for the air-blow condition than the wet condition. The signal intensities decreased during storage in air at room temperature. ConclusionTooth-substrate hydration conditions thus appeared to influence time domain-OCT imaging.

A prospective randomised study using optical coherence tomography to assess endothelial coverage and neointimal proliferation at 6-months after implantation of a coronary everolimus-eluting stent compared with a bare metal stent postdilated with a paclitaxel-eluting balloon (OCTOPUS Trial): rationale, design and methods

Safety concerns regarding use of drug-eluting stent systems (DES) are related mostly to late stent thrombosis, which is facilitated by incomplete stent endothelial coverage. Specific information about time course and amount of endothelial strut coverage of different DES is required, in order to further refine the concept of antiplatelet therapy after DES implantation. Optical coherence tomography (OCT) is emerging as a new gold standard for endovascular imaging of stents, atherosclerosis progression, vulnerable plaque, and neointimal proliferation. The aim of this study is a comparative evaluation using OCT of the XIENCE V everolimus-eluting stent (Abbot Vascular, Santa Clara, CA, USA) on one hand, and the bare metal stent Coroflex Blue postdilated with the paclitaxel-eluting balloon Sequent Please (both from B Braun Melsungen AG, Melsungen, Germany) on the other hand, with respect to endothelial coverage and neointimal proliferation. Eighty patients scheduled for elective percutaneous coronary intervention (PCI) of a native coronary stenosis suitable for DES implantation and OCT imaging are scheduled to be openly randomised 1:1 to either XIENCE or Coroflex Blue/Sequent Please. The study is conducted prospectively at a university high-volume PCI centre with OCT expertise. Angiographic follow-up and time-domain OCT imaging with motorised pull-back at 1 mm/s are planned six months after study stent implantation in all patients. OCT endpoints are: (1) endothelial coverage, expressed as % of struts without coverage and % of stent length containing non-covered struts, and respectively (2) neointimal proliferation, given as % neointimal volumetric proliferation within the whole stent and also as peak focal % neointimal area proliferation. The study is not powered for clinical endpoints, which are: subacute or late stent thrombosis and need for revascularisation of the stent segment. Given the high number of measurements (15 cross-section images / 1 mm stent length), OCT endpoints are likely to reach significance at the level p <0.05, if the drop-out rate in follow-up does not exceed 20%. The study is currently on-going and its termination is scheduled for February 2010. (ClinicalTrials.gov identifier: NCT01056744).

Comparison of retinal thickness by Fourier-domain optical coherence tomography and OCT retinal image analysis software segmentation analysis derived from Stratus optical coherence tomography images

To compare thickness measurements between Fourier-domain optical coherence tomography (FD-OCT) and time-domain OCT images analyzed with a custom-built OCT retinal image analysis software (OCTRIMA). Macular mapping (MM) by StratusOCT and MM5 and MM6 scanning protocols by an RTVue-100 FD-OCT device are performed on 11 subjects with no retinal pathology. Retinal thickness (RT) and the thickness of the ganglion cell complex (GCC) obtained with the MM6 protocol are compared for each early treatment diabetic retinopathy study (ETDRS)-like region with corresponding results obtained with OCTRIMA. RT results are compared by analysis of variance with Dunnett post hoc test, while GCC results are compared by paired t-test. A high correlation is obtained for the RT between OCTRIMA and MM5 and MM6 protocols. In all regions, the StratusOCT provide the lowest RT values (mean difference 43 ± 8 μm compared to OCTRIMA, and 42 ± 14 μm compared to RTVue MM6). All RTVue GCC measurements were significantly thicker (mean difference between 6 and 12 μm) than the GCC measurements of OCTRIMA. High correspondence of RT measurements is obtained not only for RT but also for the segmentation of intraretinal layers between FD-OCT and StratusOCT-derived OCTRIMA analysis. However, a correction factor is required to compensate for OCT-specific differences to make measurements more comparable to any available OCT device.

Three-dimensional High-speed Optical Coherence Tomography for Diagnosis of Hypotony Maculopathy After Glaucoma Filtration Surgery

To describe the clinical findings of hypotony maculopathy using 3-dimensional (3D) topography maps reconstructed from spectral domain optical coherence tomography (SD-OCT) imaging, and compare SD-OCT with time domain OCT (TD-OCT) for hypotony maculopathy diagnosis. This was an observational noncomparative case series comprising 7 patients with hypotony maculopathy after trabeculectomy with mitomycin-C. All patients underwent consecutive imaging with TD-OCT (Stratus OCT) and SD-OCT using various high-resolution instruments. 3-D surface maps obtained using SD-OCT were compared with linear scans obtained using TD-OCT. Two of 7 eyes had minimally detectable folds with TD-OCT imaging in the vertical axis. Five eyes did not show folds on TD-OCT. 3-D topographic maps using SD-OCT demonstrated advanced retinal and subretinal folds throughout the macular region in all 7 patients. All eyes (4 cases) with no topographically detectable folds using SD-OCT within the foveal avascular zone had visual acuity of 20/25 or better. Three eyes with obvious contour disruption within the foveal pit had visual acuity ranging from 20/30 to 20/70. SD-OCT using 3-D surface topography mapping provides greater sensitivity for hypotony maculopathy diagnosis and monitoring as compared with TD-OCT. Disruption of the foveal pit as detected using SD-OCT is associated with reduced visual acuity.

Imaging of subcutaneous blood vessels and flow velocity profiles by optical coherence tomography

We have applied a compact low power rapid scanning Doppler Optical Coherence Tomography system to monitor multi-dimensional velocity profiles within the complex vessels and simultaneous real-time non-invasive imaging of skin tissues morphology in vivo, in the wavelength range of 1.3–1.5 nm. Optical clearing of skin tissues has been utilized to achieve depth of OCT images up to 1.7 mm. Current approach enables applying low-power (0.4–0.5 mW) and low-noise broadband near-infrared light sources and obtaining OCT images with down to 12 μm spatial resolution. Two-dimensional time-domain OCT images of complex flow velocity profiles in blood vessel phantom and in vivo subcutaneous human skin tissues are presented. The effect of optical clearing on in vivo images is demonstrated and discussed.

In-vivo retinal imaging by optical coherence tomography using an RSOD-based phase modulator

Fourier-domain rapid scanning optical delay line (RSOD) was introduced for phase modulation and depth scanning in a time-domain optical coherence tomography (TD-OCT) system. Investigation of parameter optimization of RSOD was conducted. Experiments for RSOD characterization at different parameters of the groove pitch, focal length, galvomirror size, etc. were performed. By implementing the optimized RSOD in our established TD-OCT system with a broadband light source centered at 840 nm with 50 nm bandwidth, in vivo retina imaging of a rabbit was presented, demonstrating the feasibility of high-quality TD-OCT imaging using an RSOD-based phase modulator.

Outer retinal folds in highly myopic macular hole and retinal detachment.

To report the intraretinal microstructure in macular hole and retinal detachment in a highly myopic eye using spectral-domain optical coherence tomography. Two eyes of two patients with macular holes and retinal detachment were examined with spectral-domain optical coherence tomography. Fundus photographs and time-domain optical coherence tomography images also were obtained. Spectral-domain optical coherence tomography disclosed outer retinal fold formation toward the subretinal space at the level of the photoreceptor inner and outer segment layer, the cystoid spaces expanding outside of the macular hole and epiretinal membrane in all eyes. The outer retinal fold may indicate the tension of the inner retina and the redundancy of the outer retina, which may be responsible for the occurrence of macular holes and retinal detachment in these eyes.

Ultrahigh-resolution optical coherence tomography with a fiber laser source at 1??m

We report a compact, high-power, fiber-based source for ultrahigh-resolution optical coherence tomography (OCT) near 1 microm. The practical source is based on a short-pulse, ytterbium-doped fiber laser and on generation of a continuum spectrum in a photonic crystal fiber. The broadband emission has an average power of 140 mW and offers an axial resolution of 2.1 microm in air (<1.6 microm in biological tissue). The generation of a broad bandwidth is robust and efficient. We demonstrate ultrahigh-resolution, time-domain OCT imaging of in vitro and in vivo biological tissues.

Optical coherence tomography using a continuous-wave, high-power, Raman continuum light source

High performance, short coherence length light sources with broad bandwidths and high output powers are critical for high-speed, ultrahigh resolution OCT imaging. We demonstrate a new, high performance light source for ultrahigh resolution OCT. Bandwidths of 140 nm at 1300 nm center wavelength with high output powers of 330 mW are generated by an all-fiber Raman light source based on a continuous-wave Yb-fiber laser-pumped microstructure fiber. The light source is compact, robust, turnkey and requires no optical alignment. In vivo, ultrahigh resolution, high-speed, time domain OCT imaging with <5 microm axial resolution is demonstrated.