Five-Site Water Models for Ice and Liquid Water Generated by a Series-Parallel Machine Learning Strategy.

Icing, a common natural phenomenon, always originates from a molecule. Molecular simulation is crucial for understanding the relevant process but still faces a great challenge in obtaining a uniform and accurate description of ice and liquid water with limited model parameters. Here, we propose a series-parallel machine learning (ML) approach consisting of a classification back-propagation neural network (BPNN), parallel regression BPNNs, and a genetic algorithm to establish conventional TIP5P-BG and temperature-dependent TIP5P-BGT models. The established water models exhibit a comprehensive balance among the crucial physical properties (melting point, density, vaporization enthalpy, self-diffusion coefficient, and viscosity) with mean absolute percentage errors of 2.65 and 2.40%, respectively, and excellent predictive performance on the related properties of liquid water. For ice, the simulation results on the critical nucleus size and growth rate are in good accordance with experiments. This work offers a powerful molecular model for phase transition and icing in nanoconfinement and a construction strategy for a complex molecular model in the extreme case.

Development and evaluation of a large language model of ophthalmology in Chinese.

BACKGROUND: Large language models (LLMs), such as ChatGPT, have considerable implications for various medical applications. However, ChatGPT's training primarily draws from English-centric internet data and is not tailored explicitly to the medical domain. Thus, an ophthalmic LLM in Chinese is clinically essential for both healthcare providers and patients in mainland China. METHODS: We developed an LLM of ophthalmology (MOPH) using Chinese corpora and evaluated its performance in three clinical scenarios: ophthalmic board exams in Chinese, answering evidence-based medicine-oriented ophthalmic questions and diagnostic accuracy for clinical vignettes. Additionally, we compared MOPH's performance to that of human doctors. RESULTS: In the ophthalmic exam, MOPH's average score closely aligned with the mean score of trainees (64.7 (range 62-68) vs 66.2 (range 50-92), p=0.817), but achieving a score above 60 in all seven mock exams. In answering ophthalmic questions, MOPH demonstrated an adherence of 83.3% (25/30) of responses following Chinese guidelines (Likert scale 4-5). Only 6.7% (2/30, Likert scale 1-2) and 10% (3/30, Likert scale 3) of responses were rated as 'poor or very poor' or 'potentially misinterpretable inaccuracies' by reviewers. In diagnostic accuracy, although the rate of correct diagnosis by ophthalmologists was superior to that by MOPH (96.1% vs 81.1%, p>0.05), the difference was not statistically significant. CONCLUSION: This study demonstrated the promising performance of MOPH, a Chinese-specific ophthalmic LLM, in diverse clinical scenarios. MOPH has potential real-world applications in Chinese-language ophthalmology settings.

Generative adversarial networks synthetic optical coherence tomography images as an education tool for image diagnosis of macular diseases: a randomized trial.

Purpose: This study aimed to evaluate the effectiveness of generative adversarial networks (GANs) in creating synthetic OCT images as an educational tool for teaching image diagnosis of macular diseases to medical students and ophthalmic residents. Methods: In this randomized trial, 20 fifth-year medical students and 20 ophthalmic residents were enrolled and randomly assigned (1:1 allocation) into Group real OCT and Group GANs OCT. All participants had a pretest to assess their educational background, followed by a 30-min smartphone-based education program using GANs or real OCT images for macular disease recognition training. Two additional tests were scheduled: one 5 min after the training to assess short-term performance, and another 1 week later to assess long-term performance. Scores and time consumption were recorded and compared. After all the tests, participants completed an anonymous subjective questionnaire. Results: Group GANs OCT scores increased from 80.0 (46.0 to 85.5) to 92.0 (81.0 to 95.5) 5 min after training (p < 0.001) and 92.30 ± 5.36 1 week after training (p < 0.001). Similarly, Group real OCT scores increased from 66.00 ± 19.52 to 92.90 ± 5.71 (p < 0.001), respectively. When compared between two groups, no statistically significant difference was found in test scores, score improvements, or time consumption. After training, medical students had a significantly higher score improvement than residents (p < 0.001). Conclusion: The education tool using synthetic OCT images had a similar educational ability compared to that using real OCT images, which improved the interpretation ability of ophthalmic residents and medical students in both short-term and long-term performances. The smartphone-based educational tool could be widely promoted for educational applications.Clinical trial registration: https://www.chictr.org.cn, Chinese Clinical Trial Registry [No. ChiCTR 2100053195].

Clinical observation of a modified technique for intrascleral fixation of flanged three-piece foldable intraocular lenses through a Hoffman pocket.

Purpose: To present the outcomes of a new technique for intrascleral fixation of a flanged three-piece foldable intraocular lens (IOL) without a conjunctival incision. Materials and methods: We retrospectively reviewed a consecutive series of 12 eyes of 12 patients who underwent scleral IOL fixation using this technique. Results: The follow-up period ranged 3-12 months. There was a significant improvement in best-corrected visual acuity, from 0.8 (1.6) logarithm of the minimum angle of resolution (logMAR) preoperatively to 0.45 (0.8) logMAR at the final postoperative follow-up (p = 0.012). Notable complications included one case of pupillary IOL capture and increased intraocular pressure. Conclusion: Our novel technique is a viable solution for managing secondary IOL fixation, enabling the use of a wider variety of IOLs and simplifying the reposition process for dislocated three-piece IOLs. This approach has the potential to lower complication rates and enhance patients' recovery.

Magnetic Non-Spherical Particles Inducing Vortices in Microchannel for Effective Mixing.

Mixing in microfluidic channels is dominated by diffusion owing to the absence of chaotic flow. However, high-efficiency microscale mixing over short distances is desired for the development of lab-on-chip systems. Here, enhanced mixing in microchannels achieved using magnetic nonspherical particles (MNSPs), is reported. Benefiting from the nonspherical shape of the MNSPs, secondary vortices exhibiting cyclical characteristics appear in microchannels when the MNSPs rotate under an external magnetic field. Increasing the rotation rate enlarges the secondary vortices, expanding the mixing zone and enhancing the mixing, resulting in a mixing efficiency exceeding 0.9 at Re of 0.069-0.69. Complementary micro-particle image velocimetry (µPIV) for flow field analysis clarifies the mixing mechanism. In addition, a chaotic vortex area is generated in the presence of two MNSPs, which shortens the distance required for achieving an appropriate mixing efficiency. This study demonstrates the potential of employing MNSPs as efficient mixers in lab-on-chip devices.

Modified Trisection Technique: One-Trip Explantation for Foldable Intraocular Lens.

PURPOSE: We present a new technique that allows an intraocular lens to be explanted through the small incisions used in modern cataract surgery. METHODS AND RESULTS: The intraocular lens optic is cut into three connected pieces at the 1-mm-wide end with scissors. Then, with the stabilizing counterforce provided by a pair of vitreoretinal forceps through a paracentesis, the middle piece is removed first, followed by the two side pieces connected with haptics flipped over at the connected part. These two parts overlap each other when passing through the incision, eventually resulting in the explantation of the intraocular lens, as an intact piece. CONCLUSION: We believe this method provides a simple and effective way to remove intraocular lens through very small incisions, which could also reduce complications and hasten patient's recovery.

Inflammatory ocular events after inactivated COVID-19 vaccination.

To report potential vaccine-induced inflammatory ocular adverse events following inactivated COVID-19 vaccination. Retrospective study of patients with uveitis and other ocular complications following inactivated coronavirus disease 2019 (COVID-19) vaccination at a tertiary referral center between May 2021 and August 2021. Data collection consisted of demographic and clinical data. The study included 8 eyes of 5 patients (4 females, 1 male), with a mean age of 37.2 ± 12.5 years (range 28-59 years). Mean time between vaccination and ocular complications onset was 13.2 ± 11.9 days (range 3-30 days), including two patients after the first dose of the vaccine and 3 patients after the second dose. The cases reported were three anterior uveitis, one herpetic keratitis and iridocyclitis, and one posterior uveitis. Patients received treatment with local and/or systemic steroids and all the patients had good visual outcomes. Ocular inflammatory events may occur after vaccination with possible gender preponderance. However, they are rare and manageable. Overall, the efficacy and safety of vaccination should be emphasized.

Self-Bending Behavior and Varying Bending Stiffness of Black Phosphorus/Molybdenum Disulfide (BP/MoS2) Heterostructure.

Vertically-stacked black phosphorus/molybdenum disulfide (BP/MoS2) heterostructures have broad prospects in flexible electronics. Bending is a common and highly concerned deformation for these flexible devices. However, the discrepancy in structures and properties among the components of 2D heterostructures often induces complex bending deformations. Here, the bending behaviors of BP, MoS2 and BP/MoS2 are investigated based on a molecular dynamics simulation. Compared with the constant bending stiffness of individual BP and MoS2, that of BP/MoS2 varies with the bending angle. Notably, a self-bending configuration induced by the lattice mismatch and size difference is found in BP/MoS2. The corresponding self-bending amplitude depends on the degree of size difference of each component and the "soft/hard" competition between them. Moreover, the size difference leads to a weakened bending stiffness, which is ascribed to the reduction in interlayer interaction. A prediction formula is proposed to evaluate the bending stiffness of BP/MoS2 with the size difference. This finding reveals novel ways for regulating the bending properties of 2D heterostructures, including the bending angle, characteristic size and stacking order. It offers an effective strategy for designing flexible devices with tunable bending performance.

Controllable and Gradient Wettability of Bilayer Two-Dimensional Materials Regulated by Interlayer Distance.

Surfaces with controllable and gradient wettability often require an elaborate design of the microstructure or its response under electrical, thermal, optical, pH, and other stimuli. Generally, the wettability change under these physical or chemical effects relies on a complex mechanism that is difficult to be quantitatively described. In this study, an online controlling strategy for surface wettability and the corresponding theoretical model are put forward based on a bilayer graphene-like atomic structure. Molecular dynamics results indicate that the surface wettability varies toward hydrophilicity after sticking a bottom material regardless of its wettability. But such an influence becomes weak with increasing interlayer distance, and the overall wettability approaches that of the upper layer material gradually. This variation is elucidated by the increase of the work of adhesion, providing new insight into the wetting transparency of graphene. A theoretical model of the governing relationship is established based on the work of adhesion, which correlates the overall surface wettability with the interlayer distance and the wettabilities of individual materials. Moreover, a surface with a uniform wettability gradient is achieved by inclining the bottom material. The spontaneous and steady motion of droplets can be induced by this gradient wettability. The relevant speedup behavior is evaluated through a theoretical model considering the varying interlayer distance, which reveals the critical role of the lower layer. This study proposes a novel strategy for controllable wetting and relevant gradient surfaces using prevailing two-dimensional materials, paving new routes to many applications such as microfluidic chips, virus diagnosis, and intelligent sensors.

Generating Synthesized Ultrasound Biomicroscopy Images from Anterior Segment Optical Coherent Tomography Images by Generative Adversarial Networks for Iridociliary Assessment.

INTRODUCTION: The aim of this study was to investigate the feasibility of generating synthesized ultrasound biomicroscopy (UBM) images from swept-source anterior segment optical coherent tomography (SS-ASOCT) images using a cycle-consistent generative adversarial network framework (CycleGAN) for iridociliary assessment on a cohort presenting for primary angle-closure screening. METHODS: The CycleGAN architecture was adopted to synthesize high-resolution UBM images trained on the SS-ASOCT dataset from the department of ophthalmology, Xinhua Hospital. The performance of the CycleGAN model was further tested in two separate datasets using synthetic UBM images from two different ASOCT modalities (in-distribution and out-of-distribution). We compared the ability of glaucoma specialists to assess the image quality of real and synthetic images. UBM measurements, including anterior chamber, iridociliary parameters, were compared between real and synthetic UBM images. Intra-class correlation coefficients, coefficients of variation, and Bland-Altman plots were used to assess the level of agreement. The Fréchet Inception Distance (FID) was measured to evaluate the quality of the synthetic images. RESULTS: The whole trained dataset included anterior chamber angle images, of which 4037 were obtained by SS-ASOCT and 2206 were obtained by UBM. The image quality of real versus synthetic SS-ASOCT images was similar as assessed by two glaucoma specialists. The Bland-Altman analysis also suggested high consistency between measurements of real and synthetic UBM images. In addition, there was fair to excellent agreement between real and synthetic UBM measurements for the in-distribution dataset (ICC range 0.48-0.97) and the out-of-distribution dataset (ICC range 0.52-0.86). The FID was 21.3 and 24.1 for the synthetic UBM images from the in-distribution and out-of-distribution datasets, respectively. CONCLUSION: We developed a CycleGAN model to translate UBM images from non-contact SS-ASOCT images. The CycleGAN synthetic UBM images showed fair to excellent reproducibility when compared with real UBM images. Our results suggest that the CycleGAN technique is a promising tool to evaluate the iridociliary and anterior chamber in an alternative non-contact method.

One-Year Outcomes of Modified Technique for Scleral Fixation of a Three-Piece Intraocular Lens Without Conjunctival Opening.

Purpose: This study aimed to present the 1-year follow-up of a modified technique for scleral fixation of three-piece intraocular lens (IOLs) without conjunctival incision. Materials and Methods: A retrospective chart review of a consecutive series of 10 eyes of nine patients who underwent scleral IOL fixation using the modified technique was performed. Data were collected 1 year after surgery for all patients. Results: The range of follow-up time was between 1 year and 31 months. At the last follow-up point, the IOL was well-positioned and the visual acuity was good (as limited by primary diseases). Short-term complications included pupillary IOL capture (n = 1) and decreased intraocular pressure (n = 1), and no long-term complications were observed. Conclusion: Outcome data support this technique as a viable option for the management of secondary IOL fixation with flexible usage of more designs of IOLs.

A Novel Subfoveal Perfluorocarbon Liquid Removal Technique Combining a 25-Gauge Retrobulbar Needle With a Built-in 30-Gauge Needle.

Introduction: To report a novel combining a 25-gauge retrobulbar needle with a built-in 30-gauge needle surgical technique for subfoveal perfluorocarbon liquid (PFCL) removal. Materials and Methods: Fourteen eyes of 14 patients who underwent subfoveal PFCL removal with a 25-gauge retrobulbar needle combined with a built-in 30-gauge needle were studied. The 30-gauge needle was inserted into the 25-gauge retrobulbar needle. The bent tip of the built-in 30-gauge needle was used to create a 30-gauge retinotomy at the farthest edge of the subfoveal PFCL droplet. Then, a flute cannula was used to aspirate the PFCL through the previously created retinotomy. The best-corrected visual acuity (BCVA) was determined, previous surgical history and post-operative complications were recorded. Results: Fourteen cases were analyzed. Most eyes (92.85%) showed an improvement in BCVA after surgery. The mean change in the BCVA was -0.7 ± 0.72 logarithm of the minimum angle of resolution (logMAR) units (p = 0.006). Post-operative complications included a self-healing macular hole in one eye and vitreous hemorrhage in one eye. Post-operative optical coherence tomography confirmed removal of the subfoveal PFCL with restoration of the macular fovea. Conclusion: Combining a 25-gauge retrobulbar needle with a built-in 30-gauge needle to remove subfoveal PFCL is easy to perform and carries little potential risk of subretinal impairment. This method also provides relatively good macular contour with functional improvement.

Secondary glaucoma caused by a special type of persistent fetal vasculature.

PURPOSE: To investigate the clinical features and surgical outcomes in infants with glaucoma secondary to a special anterior-anterior type of persistent fetal vasculature (AAPFV). METHODS: This study retrospectively reviewed the medical records of infants who underwent of the synechialysis, pupilloplasty, with or without lensectomy and limbal vitrectomy due to AAPFV and with at least 6 months of postoperative follow-up. RESULTS: Eleven patients were included. The median age at surgery was 4.0 months (interquartile range: 7 months). The mean follow-up was 21.0 ± 11.3 months. All patients achieved a normal anterior chamber, improved pupillary configuration, and normal intraocular pressure (IOP), except one that developed phthisis bulbi at the last visit. A total of 81.8% (9/11) eyes exhibited improved corneal transparency. Histopathologic findings of four pupillary membranous specimens under light microscopy showed similar components compared with PFV. Two eyes developed postoperative complications, including retinal detachment and hyphema, requiring additional surgeries. Postoperative visual acuity changed from no light perception to light perception in 6/9 patients. CONCLUSIONS: AAPFV is a special type of PFV with a potential for secondary glaucoma. Surgery treatment may offer better vision with improved cosmetic outcomes and a better controlled IOP. TRIAL REGISTRATION: The study was approved by the local institutional review board (IRB) (Approval No. XHEC-D-2021-043, Ethical Committee of Xin Hua Hospital Affiliated to Shanghai Jiao Tong University, Shanghai, China).

Development and Clinical Validation of Semi-Supervised Generative Adversarial Networks for Detection of Retinal Disorders in Optical Coherence Tomography Images Using Small Dataset.

PURPOSE: To develop and test semi-supervised generative adversarial networks (GANs) that detect retinal disorders on optical coherence tomography (OCT) images using a small-labeled dataset. METHODS: From a public database, we randomly chose a small supervised dataset with 400 OCT images (100 choroidal neovascularization, 100 diabetic macular edema, 100 drusen, and 100 normal) and assigned all other OCT images to unsupervised dataset (107,912 images without labeling). We adopted a semi-supervised GAN and a supervised deep learning (DL) model for automatically detecting retinal disorders from OCT images. The performance of the 2 models was compared in 3 testing datasets with different OCT devices. The evaluation metrics included accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curves. RESULTS: The local validation dataset included 1000 images with 250 from each category. The independent clinical dataset included 366 OCT images using Cirrus OCT Shanghai Shibei Hospital and 511 OCT images using RTVue OCT from Xinhua Hospital respectively. The semi-supervised GANs classifier achieved better accuracy than supervised DL model (0.91 vs 0.86 for local cell validation dataset, 0.91 vs 0.86 in the Shanghai Shibei Hospital testing dataset, and 0.93 vs 0.92 in Xinhua Hospital testing dataset). For detecting urgent referrals (choroidal neo-vascularization and diabetic macular edema) from nonurgent referrals (drusen and normal) on OCT images, the semi-supervised GANs classifier also achieved better area under the receiver operating characteristic curves than supervised DL model (0.99 vs 0.97, 0.97 vs 0.96, and 0.99 vs 0.99, respectively). CONCLUSIONS: A semi-supervised GAN can achieve better performance than that of a supervised DL model when the labeled dataset is limited. The current study offers utility to various research and clinical studies using DL with relatively small datasets. Semi-supervised GANs can detect retinal disorders from OCT images using relatively small dataset.

Biomimetic Janus photonic soft actuator with structural color self-reporting.

Soft actuators with variable signal/color play an important role in the fields of targeted locomotion, artificial phototropism, drug screening, cargo transportation, and interactive sensing. The ability to achieve rapid response, large curvature, wide bending angle, and full-color display continues to be an unresolved challenge for artificial actuating materials. Inspired by the angle-dependent structural color of broad-tailed hummingbird and the Janus wettability of the lotus leaf, a Janus photonic soft actuator (JPSA) was fabricated by integrating an underwater super-oleophilic copper micro-nano array and oil-phobic inverse opal through a Laplace channel. The JPSA exhibits unidirectional permeability to underwater oil droplets. Attractively, with the combination of a swellable super-oleophilic surface and photonic crystals, JPSAs were endowed with oil-controlled reversible bending behavior with self-reporting angle-dependent color indication. We described for the first time the directional actuating mechanism induced by underwater oil unidirectional penetration and revealed the corresponding actuating kinetics and the inner-stress distribution/transfer by using structural color. As an extension of such theory, a rapid responsive JPSA with a wide bending angle and full-color self-reporting is further fabricated. This work provides an efficient strategy for oil directional transportation and separation in aqueous media and inspires the fabrication of a soft actuator/sensor with structural color self-reporting.

Influence of Mo Segregation at Grain Boundaries on the High Temperature Creep Behavior of Ni-Mo Alloys: An Atomistic Study.

Based on molecular dynamics simulations, the creep behaviors of nanocrystalline Ni before and after the segregation of Mo atoms at grain boundaries are comparatively investigated with the influences of external stress, grain size, temperature, and the concentration of Mo atoms taken into consideration. The results show that the creep strain rate of nanocrystalline Ni decreases significantly after the segregation of Mo atoms at grain boundaries due to the increase of the activation energy. The creep mechanisms corresponding to low, medium, and high stress states are respectively diffusion, grain boundary slip and dislocation activities based on the analysis of stress exponent and grain size exponent for both pure Ni and segregated Ni-Mo samples. Importantly, the influence of external stress and grain size on the creep strain rate of segregated Ni-Mo samples agrees well with the classical Bird-Dorn-Mukherjee model. The results also show that segregation has little effect on the creep process dominated by lattice diffusion. However, it can effectively reduce the strain rate of the creep deformation dominated by grain boundary behaviors and dislocation activities, where the creep rate decreases when increasing the concentration of Mo atoms at grain boundaries within a certain range.

Detection of shallow anterior chamber depth from two-dimensional anterior segment photographs using deep learning.

BACKGROUND: The purpose of this study was to implement and evaluate a deep learning (DL) approach for automatically detecting shallow anterior chamber depth (ACD) from two-dimensional (2D) overview anterior segment photographs. METHODS: We trained a DL model using a dataset of anterior segment photographs collected from Shanghai Aier Eye Hospital from June 2018 to December 2019. A Pentacam HR system was used to capture a 2D overview eye image and measure the ACD. Shallow ACD was defined as ACD less than 2.4 mm. The DL model was evaluated by a five-fold cross-validation test in a hold-out testing dataset. We also evaluated the DL model by testing it against two glaucoma specialists. The performance of the DL model was calculated by metrics, including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). RESULTS: A total of 3753 photographs (1720 shallow AC and 2033 deep AC images) were assigned to the training dataset, and 1302 photographs (509 shallow AC and 793 deep AC images) were held out for two internal testing datasets. In detecting shallow ACD in the internal hold-out testing dataset, the DL model achieved an AUC of 0.86 (95% CI, 0.83-0.90) with 80% sensitivity and 79% specificity. In the same testing dataset, the DL model also achieved better performance than the two glaucoma specialists (accuracy of 80% vs. accuracy of 74 and 69%). CONCLUSIONS: We proposed a high-performing DL model to automatically detect shallow ACD from overview anterior segment photographs. Our DL model has potential applications in detecting and monitoring shallow ACD in the real world. TRIAL REGISTRATION: http://clinicaltrials.gov , NCT04340635 , retrospectively registered on 29 March 2020.

Machine learning for reparameterization of four-site water models: TIP4P-BG and TIP4P-BGT.

Parameterizing an effective water model is a challenging issue because of the difficulty in maintaining a comprehensive balance among the diverse physical properties of water with a limited number of parameters. The advancement in machine learning provides a promising path to search for a reliable set of parameters. Based on the TIP4P water model, hence, about 6000 molecular dynamics (MD) simulations for pure water at 1 atm and in the range of 273-373 K are conducted here as the training data. The back-propagation (BP) neural network is then utilized to construct an efficient mapping between the model parameters and four crucial physical properties of water, including the density, vaporization enthalpy, self-diffusion coefficient and viscosity. Without additional time-consuming MD simulations, this mapping operation could result in sufficient and accurate data for high-population genetic algorithm (GA) to optimize the model parameters as much as possible. Based on the proposed parameterizing strategy, TIP4P-BG (a conventional four-site water model) and TIP4P-BGT (an advanced model with temperature-dependent parameters) are established. Both the water models exhibit excellent performance with a reasonable balance among the four crucial physical properties. The relevant mean absolute percentage errors are 3.53% and 3.08%, respectively. Further calculations on the temperature of maximum density, isothermal compressibility, thermal expansion coefficient, radial distribution function and surface tension are also performed and the resulting values are in good agreement with the experimental values. Through this water modeling example, the potential of the proposed data-driven machine learning procedure has been demonstrated for parameterizing a MD-based material model.