Comparison of corneal thickness measurements using ultrasound pachymetry, noncontact tonopachy, Pentacam HR, and Fourier-domain OCT.

ABSTRACT: We retrospectively compared the central corneal thickness (CCT) obtained by ultrasound pachymetry (USP; SP-3000, Tomey Corp., Nagoya, Japan), non-contact tonopachy (TP) (NT-530P, Nidek Co., Ltd., Gamagori, Japan), Pentacam HR (OCULUS Inc., Wetzlar, Germany), and RTVue optical coherence tomography (OCT) (Optovue Inc., Fremont, CA, USA) in 78 eyes of 78 healthy subjects with myopia. Agreement between the measurement methods was evaluated using 95% confidence intervals for the limits of agreement (LoA). The mean CCT values were 546.9 ± 34.7, 548.1 ±â€Š33.5, 559.2 ±â€Š34.0, and 547.2 ±â€Š34.8 µm for USP, non-contact TP, Pentacam, and RTVue, respectively. The thickest and the thinnest mean CCT values corresponded to those obtained by Pentacam HR and USP, respectively. Plots of the differences against the means showed the best agreement between USP and RTVue (LoA, 10.14-10.70 µm), while the largest discrepancy was observed between RTVue and Pentacam systems (LoA, -25.47-1.44 µm). Our data showed that CCT measurements using these 4 instruments were well correlated. However, the results from Pentacam differed significantly from those of the other instruments.

Prediction of Phakic Intraocular Lens Vault Using Machine Learning of Anterior Segment Optical Coherence Tomography Metrics.

PURPOSE: To compare the achieved vault using the conventional manufacturer's nomogram and the predicted vault using machine learning, in a large cohort of eyes undergoing posterior chamber phakic intraocular lens (EVO implantable collamer lens [ICL]; STAAR Surgical) implantation. DESIGN: Reliability and validity assessment and comparison. METHODS: Our study comprised a total of 1,745 eyes of 1,745 consecutive patients (mean age ± standard deviation, 26.2 ± 6.8 years) undergoing ICL implantation. At 1 month postoperatively, we quantitatively measured the actual vault, and compared it with the predicted vault using machine leaning of anterior segment optical coherence tomography metrics. RESULTS: All machine learning methods, such as support vector regressor (SVR), gradient boost regressor (GBR), random forest regressor (RFR), and linear regressor (LR), showed significantly less mean absolute error and higher percentages of eyes within 50-200 µm of the targeted ICL vault than the conventional nomogram (P < .001). The RFR provided the lowest mean absolute errors and the highest percentages of eyes within 50 to 200 µm, followed by the GBR, the LR, and the SVR (P < .01). CONCLUSIONS: Machine learning of the preoperative anterior segment optical coherence tomography metrics, especially the RFR, provided significantly higher predictability of the ICL vault than the conventional manufacturer's nomogram, suggesting that it will become an aid for predicting the ICL vault and subsequently selecting the proper ICL size in daily practice.

Deep learning can generate traditional retinal fundus photographs using ultra-widefield images via generative adversarial networks.

BACKGROUND AND OBJECTIVE: Retinal imaging has two major modalities, traditional fundus photography (TFP) and ultra-widefield fundus photography (UWFP). This study demonstrates the feasibility of a state-of-the-art deep learning-based domain transfer from UWFP to TFP. METHODS: A cycle-consistent generative adversarial network (CycleGAN) was used to automatically translate the UWFP to the TFP domain. The model was based on an unpaired dataset including anonymized 451 UWFP and 745 TFP images. To apply CycleGAN to an independent dataset, we randomly divided the data into training (90%) and test (10%) datasets. After automated image registration and masking dark frames, the generator and discriminator networks were trained. Additional twelve publicly available paired TFP and UWFP images were used to calculate the intensity histograms and structural similarity (SSIM) indices. RESULTS: We observed that all UWFP images were successfully translated into TFP-style images by CycleGAN, and the main structural information of the retina and optic nerve was retained. The model did not generate fake features in the output images. Average histograms demonstrated that the intensity distribution of the generated output images provided a good match to the ground truth images, with an average SSIM level of 0.802. CONCLUSIONS: Our approach enables automated synthesis of TFP images directly from UWFP without a manual pre-conditioning process. The generated TFP images might be useful for clinicians in investigating posterior pole and for researchers in integrating TFP and UWFP databases. This is also likely to save scan time and will be more cost-effective for patients by avoiding additional examinations for an accurate diagnosis.

Explainable Machine Learning Approach as a Tool to Understand Factors Used to Select the Refractive Surgery Technique on the Expert Level.

Purpose: Recently, laser refractive surgery options, including laser epithelial keratomileusis, laser in situ keratomileusis, and small incision lenticule extraction, successfully improved patients' quality of life. Evidence-based recommendation for an optimal surgery technique is valuable in increasing patient satisfaction. We developed an interpretable multiclass machine learning model that selects the laser surgery option on the expert level. Methods: A multiclass XGBoost model was constructed to classify patients into four categories including laser epithelial keratomileusis, laser in situ keratomileusis, small incision lenticule extraction, and contraindication groups. The analysis included 18,480 subjects who intended to undergo refractive surgery at the B&VIIT Eye center. Training (n = 10,561) and internal validation (n = 2640) were performed using subjects who visited between 2016 and 2017. The model was trained based on clinical decisions of highly experienced experts and ophthalmic measurements. External validation (n = 5279) was conducted using subjects who visited in 2018. The SHapley Additive ex-Planations technique was adopted to explain the output of the XGBoost model. Results: The multiclass XGBoost model exhibited an accuracy of 81.0% and 78.9% when tested on the internal and external validation datasets, respectively. The SHapley Additive ex-Planations explanations for the results were consistent with prior knowledge from ophthalmologists. The explanation from one-versus-one and one-versus-rest XGBoost classifiers was effective for easily understanding users in the multicategorical classification problem. Conclusions: This study suggests an expert-level multiclass machine learning model for selecting the refractive surgery for patients. It also provided a clinical understanding in a multiclass problem based on an explainable artificial intelligence technique. Translational Relevance: Explainable machine learning exhibits a promising future for increasing the practical use of artificial intelligence in ophthalmic clinics.

Deep learning-based smart speaker to confirm surgical sites for cataract surgeries: A pilot study.

Wrong-site surgeries can occur due to the absence of an appropriate surgical time-out. However, during a time-out, surgical participants are unable to review the patient's charts due to their aseptic hands. To improve the conditions in surgical time-outs, we introduce a deep learning-based smart speaker to confirm the surgical information prior to cataract surgeries. This pilot study utilized the publicly available audio vocabulary dataset and recorded audio data published by the authors. The audio clips of the target words, such as left, right, cataract, phacoemulsification, and intraocular lens, were selected to determine and confirm surgical information in the time-out speech. A deep convolutional neural network model was trained and implemented in the smart speaker that was developed using a mini development board and commercial speakerphone. To validate our model in the consecutive speeches during time-outs, we generated 200 time-out speeches for cataract surgeries by randomly selecting the surgical statuses of the surgical participants. After the training process, the deep learning model achieved an accuracy of 96.3% for the validation dataset of short-word audio clips. Our deep learning-based smart speaker achieved an accuracy of 93.5% for the 200 time-out speeches. The surgical and procedural accuracy was 100%. Additionally, on validating the deep learning model by using web-generated time-out speeches and video clips for general surgery, the model exhibited a robust and good performance. In this pilot study, the proposed deep learning-based smart speaker was able to successfully confirm the surgical information during the time-out speech. Future studies should focus on collecting real-world time-out data and automatically connecting the device to electronic health records. Adopting smart speaker-assisted time-out phases will improve the patients' safety during cataract surgeries, particularly in relation to wrong-site surgeries.

The Association between Macular Thickness and Axial Length in Myopic Eyes.

PURPOSE: To investigate the relationship between macular thickness and axial length (AL) in myopic eyes. METHODS: We included 441 myopic eyes in this study and measured macular thickness at the fovea and in other macular regions, using optical coherence tomography. We got thickness difference indices (TDIs) which by definition are the values of thickness difference obtained by subtracting the foveal thickness from that of each macula sector to evaluate macular contour. We then analyzed the relationships between AL and foveal thickness and AL and the TDIs of each macular sector. RESULTS: In polynomial regression analyses, foveal thickness slope was relatively flat up to an AL of 25.5 mm and began to rise from 25.5-26.0 mm. The TDIs were also relatively flat up to AL of 25.5mm and started to show steepened negative slopes from around AL of 25.5 mm. When grouping myopia participants as high myopia or non-high myopia based on AL of 25.5mm, all macular indices of the high myopia group showed significant correlation with AL (all p values <0.01). But all indices of non-high myopia group had no significant correlation with AL. CONCLUSIONS: Average macular thickness profiles showed that appreciable changes started at an AL of 25.5mm.

Adopting machine learning to automatically identify candidate patients for corneal refractive surgery.

Recently, it has become more important to screen candidates that undergo corneal refractive surgery to prevent complications. Until now, there is still no definitive screening method to confront the possibility of a misdiagnosis. We evaluate the possibilities of machine learning as a clinical decision support to determine the suitability to corneal refractive surgery. A machine learning architecture was built with the aim of identifying candidates combining the large multi-instrument data from patients and clinical decisions of highly experienced experts. Five heterogeneous algorithms were used to predict candidates for surgery. Subsequently, an ensemble classifier was developed to improve the performance. Training (10,561 subjects) and internal validation (2640 subjects) were conducted using subjects who had visited between 2016 and 2017. External validation (5279 subjects) was performed using subjects who had visited in 2018. The best model, i.e., the ensemble classifier, had a high prediction performance with the area under the receiver operating characteristic curves of 0.983 (95% CI, 0.977-0.987) and 0.972 (95% CI, 0.967-0.976) when tested in the internal and external validation set, respectively. The machine learning models were statistically superior to classic methods including the percentage of tissue ablated and the Randleman ectatic score. Our model was able to correctly reclassify a patient with postoperative ectasia as an ectasia-risk group. Machine learning algorithms using a wide range of preoperative information achieved a comparable performance to screen candidates for corneal refractive surgery. An automated machine learning analysis of preoperative data can provide a safe and reliable clinical decision for refractive surgery.

Asymmetry of the macular structure is associated with ocular dominance.

OBJECTIVE: We hypothesized that the ocular dominance associated with brain asymmetry would influence macular structure. We assessed the structural asymmetry of the macula by ocular dominance. METHODS: A total of 144 patients with no ophthalmic abnormalities were examined. After classifying eyes as dominant or non-dominant using the hole-in-a-card test, the macular ganglion cell inner plexiform layer and the macular thickness were measured in both groups using spectral domain optical coherence tomography, and we sought associations between ocular dominance and retinal thickness. RESULTS: Ocular dominance was more frequently found in right eyes than in left eyes (68.75% vs 31.25%, respectively; p < 0.001). Using conditioned logistic regression, the right eye and the temporal-to-nasal retinal thickness ratio of the macular ganglion cell inner plexiform layer were selected as a final model for the determination of ocular dominance (p < 0.001, p = 0.013 respectively). CONCLUSIONS: Macular vertical asymmetry was associated with ocular dominance. The macular structure and ocular dominance associated with visual cortex lateralization have functional and structural relationships.

Retrospective observational study of micro-monovision small incision lenticule extraction (SMILE) for the correction of presbyopia and myopia.

This study was aimed to evaluate refractive and visual outcomes after micro-monovision small incision lenticule extraction (SMILE) in patients with presbyopia and myopia. In total, 72 patients (144 eyes) with a mean age of 46.0 ±â€Š4.9 years were included in this study. The dominant eye was treated for distance vision and the nondominant eye for near vision by targeting between -0.50 and -1.75 diopters (D). Treatment efficacy, safety, and refractive stability were calculated from postoperative data including refraction, binocular uncorrected distance visual acuity (UDVA), binocular uncorrected near visual acuity, monocular uncorrected distance visual acuity, and monocular corrected distance visual acuity (CDVA). Six months post-surgery, binocular UDVA was better than or equal to 20/20 in 88% of patients. No loss in 2 or more lines was observed in the Snellen lines of corrected distance visual acuity. Mean spherical equivalent (SE) for the distance eye was -0.18 ±â€Š0.37 D, whereas the attempted and achieved SE in the near eye were -0.90 ±â€Š0.44 and -0.99 ±â€Š0.54 D, respectively. In total, 79% of eyes were within ±0.50 D, and 98% within ±1.00 D, of the intended refraction. A UDVA of 0.0 logMAR (20/20) or better, and an uncorrected near visual acuity of Jaeger (J) of 3 (20/32) or better, were observed in 83% of patients. Micromonovision refractive surgery using SMILE enhanced functional near vision in presbyopic patients.

Ganglion Cell-Inner Plexiform Layer, Peripapillary Retinal Nerve Fiber Layer, and Macular Thickness in Eyes with Myopic ß-Zone Parapapillary Atrophy.

Purpose. To assess the correlations of myopic ß-zone parapapillary atrophy (ß-PPA) with the optic nerve head (ONH) and retina. Methods. We selected 27 myopic patients who showed prominent ß-PPA in one eye and no ß-PPA in the other eye. We studied their macula, macular ganglion cell-inner plexiform layer (mGCIPL), peripapillary retinal nerve fiber layer (pRNFL) thickness, and ONH parameters using optical coherence tomography. Results. The average of five out of six sectors and minimum values of mGCIPL thicknesses in eyes with prominent ß-PPA discs were significantly less than those of the control eyes. The results of clock-hour sector analyses showed significant differences for pRNFL thickness in one sector. In the ONH analyses, no significant difference was observed between myopic ß-PPA and control eyes. The macular thickness of the ß-PPA eyes was thinner than control eyes in all sectors. There was a significant difference between the two groups in three sectors (the inner superior macula, inner temporal macula, and inner inferior macula) but there was no significant difference in the other sectors, including the fovea. Conclusions. The myopic ß-PPA eyes showed thinner mGCIPL, parafovea, and partial pRNFL layers compared with myopic eyes without ß-PPA.

Ocular Dominance Is Associated with the Ganglion Cell-Inner Plexiform Layer Thickness Profile in the Macula.

PURPOSE: To investigate the characteristics of macular ganglion cell-inner plexiform layer (GCIPL) thickness profiles associated with ocular dominance. SETTING: Private practice, Seoul, Republic of Korea. DESIGN: Comparative case-control study. METHODS: Both eyes of 199 participants with no ophthalmic abnormalities were included. Participants were imaged by spectral-domain optical coherence tomography, and underwent dominant eye testing using a hole-in-a-card test (sighting dominance) at the same visit. Macular GCIPL, as well as circumpapillary retinal nerve fiber layer (RNFL) thickness were compared for individual patients, according to ocular dominance. RESULTS: Ocular dominance occurred predominantly in the right eye (right vs. left: 72.36 vs. 27.60%; P < 0.001). In the comparison of macular GCIPL thickness, the average (81.27±5.01 µm vs. 80.66±6.31 µm in dominant vs. non-dominant eyes), inferonasal (81.39±5.47µm vs. 80.33±6.82µm, and inferior sectors (77.95±6.05µm vs. 76.97±8.15µm) were significantly different between dominant and non-dominant eyes (P = 0.040, 0.005, and 0.032, respectively). Significant predictors of average GCIPL thickness were spherical equivalent (ß = 1.37, P<0.001), astigmatic power (ß = 1.44, P = 0.009), disc area (ß = 3.90, P < 0.001), average RNFL thickness (ß = 0.22, P<0.001), average cup-to-disc ratio (ß = 5.74, P = 0.002), difference between the inferior and superior quadrant RNFL thicknesses (ß = 0.08, P = 0.024), and ocular dominance (ß = 2.10, P = 0.020). On multivariate regression analysis, ocular dominance was correlated with average GCIPL thickness after adjusting for potential confounders (ß = 1.63, P = 0.048). CONCLUSIONS: Dominant eyes accompanied significantly thicker average macular GCIPL. This information suggests that macular GCIPL thickness may provide an indicator of the relative dominance of an eye.

Electrochemical Characteristics of Pyrolytic Carbon Coated Silicon for Anode Materials of Lithium Ion Batteries.

Pyrolytic carbon coated silicon is prepared and employ it as an anode material for lithium secondary batteries. The pyrolytic carbon coating of silicon with sucrose precursor not only provides a suitable carbon matrix but also suppresses the breaking away of Si from the current collector during the insertion and extraction of Li+. The increase of disordered carbon content leads to the increase of discharge capacity retention. The improvements of cycle stability are attributed to a decrease in the volume change of the silicon, good networking between Si particles, and good adhesion of the current collector with the active material. The nano Si combined with the disordered carbon leads to an increase of capacity retention and initial coulombic efficiency.

3D Woven-Like Carbon Micropattern Decorated with Silicon Nanoparticles for Use in Lithium-Ion Batteries.

Carbon/silicon composite materials are a promising anode substrate for use in lithium-ion batteries. In this study, we suggest a new architecture for a composite electrode made of a woven-like carbon material decorated with silicon nanoparticles. The 3D woven-like carbon (WLC) structure was fabricated using direct carbonization of multi-beam interference lithography polymer patterns. Subsequent solution coating was applied to decorate the WLC with silicon nanoparticles (SiNPs). The SiNP/WLC electrode exhibited a specific capacity of 930 mAh g(-1) , which is three times higher than the specific capacity of the bare electrode. Specifically, the SiNP/WLC electrode exhibited an outstanding retention capacity of 81 % after 50 cycles and a Coulombic efficiency of more than 98 %. This rate capability performance was attributed to the WLC structure and the uniform decoration of the SiNPs.

Si/Ti2O3/Reduced Graphene Oxide Nanocomposite Anodes for Lithium-Ion Batteries with Highly Enhanced Cyclic Stability.

Silicon (Si) has attracted tremendous attention as a high-capacity anode material for next generation Li-ion batteries (LIBs); unfortunately, it suffers from poor cyclic stability due to excessive volume expansion and reduced electrical conductivity after repeated cycles. To circumvent these issues, we propose that Si can be complexed with electrically conductive Ti2O3 to significantly enhance the reversible capacity and cyclic stability of Si-based anodes. We prepared a ternary nanocomposite of Si/Ti2O3/reduced graphene oxide (rGO) using mechanical blending and subsequent thermal reduction of the Si, TiO2 nanoparticles, and rGO nanosheets. As a result, the obtained ternary nanocomposite exhibited a specific capacity of 985 mAh/g and a Coulombic efficiency of 98.4% after 100 cycles at a current density of 100 mA/g. Furthermore, these ternary nanocomposite anodes exhibited outstanding rate capability characteristics, even with an increased current density of 10 A/g. This excellent electrochemical performance can be ascribed to the improved electron and ion transport provided by the Ti2O3 phase within the Si domains and the structurally reinforced conductive framework comprised of the rGO nanosheets. Therefore, it is expected that our approach can also be applied to other anode materials to enable large reversible capacity, excellent cyclic stability, and good rate capability for high-performance LIBs.

Preparation of Gold Nanoparticles Deposited Silicon Thin Film Electrode by Self-Assembly Method for the Employment of an Anode Material for Lithium Secondary Batteries.

This work describes a self-assembly method of gold nanoparticles coating on the surface of silicon thin films for the anode material of lithium secondary batteries. The preparation of the silicon thin films was carried out by electron cyclotron resonance metal organic chemical vapor deposition (ECR-MOCVD) process. The obtained films were further coated with (3-aminopropyl)-trimethoxysilane (APTMS) which has a role to bind the oxygen functional groups on Si surface and the gold nanoparticles. The dispersed gold nanoparticles on the surface of silicon thin films could be prepared due to self-assembly phenomena which interact between attraction and repulsion in gold nanoparticles colloidal solution (GNCS). The use of reducing agent of sodium citrate and tannic acid in GNCS significantly affected the size of gold nanoparticle in our experimental range. Based on our experimental results, the higher reversible capacity was exhibited for the silicon that was immersed in the GNCS consisted of only sodium citrate. The GNCS consisted of both sodium citrate and tannic acid produced severe coagulated nanoparticles when deposited on the silicon surface and thus inhibited the lithium movement from electrolyte to silicon surface. Consequently, the reversible capacity of silicon anode material with coagulated gold nanoparticles coating showed the reduced performance.

The foveal position relative to the optic disc and the retinal nerve fiber layer thickness profile in myopia.

PURPOSE: To evaluate retinal nerve fiber layer (RNFL) thickness profiles according to the foveal position relative to the optic disc in myopia METHODS: In 164 eyes of 164 healthy myopic subjects, the disc-foveal angle was defined as the angle between a horizontal line through the disc center and the line connecting the fovea and disc center in fundus photographs overlaid on Cirrus-HD optical coherence tomography (OCT) images. The quadrant/clock-hour based peripapillary RNFL thickness and differences between the inferior and superior (I-S) quadrant RNFL thicknesses were measured with OCT. RNFL thickness profiles were determined according to the disc-foveal angle and axial length (AL). RESULTS: As the disc-foveal angle increased (i.e., the fovea becomes more inferior to the optic disc), the superior RNFL decreased significantly (P = 0.003), whereas the inferior RNFL and I-S difference increased (P = 0.010 and P < 0.001, respectively). As the AL increased, the average and temporal RNFLs increased significantly (P = 0.013 and P < 0.001, respectively), and I-S difference was not affected (P = 0.231). The disc-foveal angle was significantly decreased with the distance between the fovea and the optic disc (P = 0.033). In multiple linear regression analysis, the disc-foveal angle was found to be a significant factor related to I-S differences, superior and inferior RNFL (all, P < 0.05) after adjusting for age, disc area, and AL. CONCLUSIONS: The intrinsic foveal position relative to the optic disc was an essential determinant of normal RNFL thickness in myopia. In particular, it was associated with the vertical asymmetry of RNFL distribution.

Si-Mn/reduced graphene oxide nanocomposite anodes with enhanced capacity and stability for lithium-ion batteries.

Although Si is a promising high-capacity anode material for Li-ion batteries (LIB), it suffers from capacity fading due to excessively large volumetric changes upon Li insertion. Nanocarbon materials have been used to enhance the cyclic stability of LIB anodes, but they have an inherently low specific capacity. To address these issues, we present a novel ternary nanocomposite of Si, Mn, and reduced graphene oxide (rGO) for LIB anodes, in which the Si-Mn alloy offers high capacity characteristics and embedded rGO nanosheets confer structural stability. Si-Mn/rGO ternary nanocomposites were synthesized by mechanical complexation and subsequent thermal reduction of mixtures of Si nanoparticles, MnO2 nanorods, and rGO nanosheets. Resulting ternary nanocomposite anodes displayed a specific capacity of 600 mAh/g with ∼90% capacity retention after 50 cycles at a current density of 100 mA/g. The enhanced performance is attributed to facilitated Li-ion reactions with the MnSi alloy phase and the formation of a structurally reinforced electroconductive matrix of rGO nanosheets. The ternary nanocomposite design paradigm presented in this study can be exploited for the development of high-capacity and long-life anode materials for versatile LIB applications.

Retinal nerve fiber layer thickness profiles associated with ocular laterality and dominance.

Although human anatomy is arranged symmetrically based on a central vertical axis, the majority of persons will use one side of their body more readily than the other. Interestingly, these lateral body dominances including ocular dominance are all rightward. The asymmetry in retinal nerve fiber layer (RNFL) thickness between the right and left eyes in healthy subjects has been reported in several studies, and the reason for this structural difference between right and left eyes is unclear. In the manuscript, we hypothesized that the characteristics of ocular dominance are reflected in the RNFL profile and may be related to inter-ocular structural differences between right and left eyes. In this study, ocular dominance occurred mostly in right eyes (right vs. left: 78.77% vs. 21.22%; P<0.001). According to ocular dominance and laterality, different relationships between the inferior and superior RNFLs were observed. The right eyes had a thicker RNFL, except in the superior quadrant, than the left eyes. Regardless of laterality, inferior RNFL was thicker than superior RNFL in the dominant eyes. To our knowledge, this paper is the first report demonstrating the RNFL characteristics associated with ocular dominance.

Effect of hydrogen plasma pretreatment on the growth of silicon nanowires and their employment as the anode material of lithium secondary batteries.

Silicon nanowires were grown from a silane and argon gas mixture directly on a stainless steel substrate by radio-frequency plasma enhanced chemical vapor deposition (RF-PECVD) and used without any further treatment as the anode in the fabrication of lithium ion batteries. It was found that suitable pretreatment of the stainless steel substrate was required for the satisfactory growth of the silicon nanowires. In this study, the substrates were polished, etched in HF solution, coated with an aluminum catalyst layer with a thickness of c.a. 10 nm and then treated with a hydrogen plasma before the growth of the silicon nanowires. SEM (Scanning Electron Microscopy) and AFM (Atomic Force Microscopy) analyses showed that the grain size and surface roughness were increased after the hydrogen plasma pretreatment. The electrochemical performance of the silicon nanowires anode was also improved when the aluminum coated stainless steel substrate was exposed to the plasma for 20 min or longer; the initial coulombic efficiency was increased from 69.7% to 82% at a current density of 30 mA cm(-2).

Diagnosis and treatment of postpartum Candida endophthalmitis.

The occurrence of endogenous Candida endophthalmitis is rare in immunocompetent patients. We report a case of a 29-year-old healthy woman who developed endogenous Candida endophthalmitis after normal spontaneous vaginal delivery. The patient was diagnosed with Candida endophthalmitis by positive vitreous culture obtained by diagnostic vitrectomy. After vitrectomy, she was treated with systemic and intravitreal antifungal therapy, and subsequently, her visual acuity recovered to 20/20 without other complications. Normal spontaneous vaginal delivery may cause endogenous Candida endophthalmitis in young healthy women. Diagnostic vitrectomy with systemic and intravitreal antifungal treatment can be useful in diagnosis and treatment of Candida endophthalmitis.