Optical coherence tomography choroidal enhancement using generative deep learning.

Spectral-domain optical coherence tomography (SDOCT) is the gold standard of imaging the eye in clinics. Penetration depth with such devices is, however, limited and visualization of the choroid, which is essential for diagnosing chorioretinal disease, remains limited. Whereas swept-source OCT (SSOCT) devices allow for visualization of the choroid these instruments are expensive and availability in praxis is limited. We present an artificial intelligence (AI)-based solution to enhance the visualization of the choroid in OCT scans and allow for quantitative measurements of choroidal metrics using generative deep learning (DL). Synthetically enhanced SDOCT B-scans with improved choroidal visibility were generated, leveraging matching images to learn deep anatomical features during the training. Using a single-center tertiary eye care institution cohort comprising a total of 362 SDOCT-SSOCT paired subjects, we trained our model with 150,784 images from 410 healthy, 192 glaucoma, and 133 diabetic retinopathy eyes. An independent external test dataset of 37,376 images from 146 eyes was deployed to assess the authenticity and quality of the synthetically enhanced SDOCT images. Experts' ability to differentiate real versus synthetic images was poor (47.5% accuracy). Measurements of choroidal thickness, area, volume, and vascularity index, from the reference SSOCT and synthetically enhanced SDOCT, showed high Pearson's correlations of 0.97 [95% CI: 0.96-0.98], 0.97 [0.95-0.98], 0.95 [0.92-0.98], and 0.87 [0.83-0.91], with intra-class correlation values of 0.99 [0.98-0.99], 0.98 [0.98-0.99], and 0.95 [0.96-0.98], 0.93 [0.91-0.95], respectively. Thus, our DL generative model successfully generated realistic enhanced SDOCT data that is indistinguishable from SSOCT images providing improved visualization of the choroid. This technology enabled accurate measurements of choroidal metrics previously limited by the imaging depth constraints of SDOCT. The findings open new possibilities for utilizing affordable SDOCT devices in studying the choroid in both healthy and pathological conditions.

Machine learning predicted inelasticity in defective two-dimensional transition metal dichalcogenides using SHAP analysis.

The manipulation of crystallographic defects in 2H-transition metal dichalcogenides (2H-TMDCs), whether pre- or post-synthesis, has garnered significant interest recently, as it holds the promise of tuning the thermal, chemical, and electronic properties of these materials. However, such desirable improvements often come at the cost of deteriorated elastic and inelastic properties, which may lead to serious concerns considering mechanical reliability issues. Therefore, persistent efforts are needed to explore the effects of energetically favorable vacancies on the mechanical properties of 2D TMDCs for an effective tuning of material properties for versatile applications. In this context, machine learning models trained on data based on molecular models can not only provide fast, efficient material models but also unearth crucial structure-property relations. However, such efforts are at an early stage of development. In this study, machine learning and deep learning techniques are used to analyze the mechanical properties of 2D transition metal dichalcogenides (TMDCs) in both pristine and defect forms. The goal is to predict failure stress, strain to failure, and strength based on chirality and strain. Various crystallographic defects were considered, and extensive molecular dynamics simulations were performed. XGBoost and densely connected neural network (DenseNet) algorithms were used to make accurate state-of-the-art predictions, and comparative evaluation and Shapley value analysis of both models are presented to improve interpretability.

Empowering the Water Oxidation Activity of the Bimetallic Metal-Organic Framework by Annexing Gold Nanoparticles over the Catalytic Surface.

Electrocatalytic water splitting to an anodic oxygen evolution reaction (OER) and a cathodic hydrogen evolution reaction (HER) is believed to be the most important application for sustainable hydrogen generation. Being a four-electron, four-proton transfer process, the OER plays the main obstacle for the same. Therefore, designing an effective electrocatalyst to minimize the activation energy barrier for the OER is a research topic of prime importance. The metal-organic framework (MOF) with a highly porous network is considered an appropriate candidate for the OER in alkaline conditions. Apart from several MOFs, the bimetallic one has an advantageous electrocatalytic performance due to the synergistic electronic interaction between two metal ions. However, most bimetallic MOFs have an obstacle to electrocatalytic application due to their low conductive nature, and therefore, they possess a barrier for charge transfer kinetics at the interface. Surface functionalization via various nanoparticles (NPs) is believed to be the most effective strategy for nullifying the conductive issue. In this work, we have designed a CoNi-based bimetallic MOF that was surface-functionalized by Au NPs (Au@CoNi-Bpy-BTC) for the OER under alkaline conditions. Au@CoNi-Bpy-BTC required an overpotential of just 330 mV, which is 56 mV lower as compared to the pristine MOF. Impedance analysis confirms an improved conductivity and charge transfer at the interface, where Au@CoNi-Bpy-BTC possesses a lower Rct value than CoNi-Bpy-BTC materials. Moreover, the Au-decorated MOF shows an 8.5 times increase in the TOF value compared to the pristine MOF. Therefore, this noble strategy toward the surface functionalization of MOFs via noble metal NPs is believed to be the most effective strategy for developing effective electrocatalysts for electrocatalytic application in energy-related fields. Overall, this report displays an exceptional correlation between the decorated NPs over the MOF surface, which can regulate the OER activity, as confirmed by experimental analysis.

Innovative Fabrication of Metal Alloy Structures via Laser-Induced Forward Transfer on Flexible Substrates.

Laser-induced forward transfer is a contactless, nozzle free process which enables accurate, precise and fast development of 3D structures. However, a number of shortcomings such as shockwave generation, poor adhesion to receiver substrates and uniform depositions limit LIFT to be utilized. Therefore, this research tends to put forward easy and effective solutions for successful mitigation of the LIFT limitations. Receiver surface modifications and low-pressure conditions are introduced through laser surface texturing (LST) and a vacuum pump. A number of textures and orientations are investigated for determining the optimal copper (Cu) deposition. Furthermore, utilization of the same laser system for LST enables the manufacturing process cost and time effective. In addition to Cu depositions, additive layers of silver (Ag) and platinum (Pt) are deposited. Finally, Ag and Pt micropillars are fabricated on their respective additive layers leading to formation of Cu-Ag and Cu-Pt alloys structure. Subsequently, electrical and material characterizations are made to validate for potential applications. Experimental evidence shows greater adhesion with electrical properties for LST-based LIFT in low pressure conditions. Finally, an energy analysis is performed based on theoretical and finite element methods (FEM) to gain greater insights into mechanics of the LIFT process.

Regulating Surface Charge by Embedding Ru Nanoparticles over 2D Hydroxides toward Water Oxidation.

Exploring highly active and earth-abundant electrocatalysts for the oxygen evolution reaction (OER) is considered one of the prime prerequisites for generating green hydrogen. Herein, a competent microwave-assisted decoration of Ru nanoparticles (NPs) over the bimetallic layered double hydroxide (LDH) material is proposed. The same has been used as an OER catalyst in a 1 M KOH solution. The catalyst shows an interesting Ru NP loading dependency toward the OER, and a concentration-dependent volcanic relationship between electronic charge and thermoneutral current densities has been observed. This volcanic relation shows that with an optimum concentration of Ru NPs, the catalyst could effectively catalyze the OER by obeying the Sabatier principle of ion adsorption. The optimized Ru@CoFe-LDH(3%) demands an overpotential value of only 249 mV to drive a current density value of 10 mA/cm2 with the highest TOF value of 14.4 s-1 as compared to similar CoFe-LDH-based materials. In situ impedance experiments and DFT studies demonstrated that incorporating the Ru NPs boosts the intrinsic OER activity of the CoFe-LDH on account of sufficient activated redox reactivities for both Co and lattice oxygen of the CoFe-LDH. As a result, compared with the pristine CoFe-LDH, the current density of Ru@CoFe-LDH(3%) at 1.55 V vs RHE normalized by ECSA increased by 86.58%. First-principles DFT analysis shows that the optimized Ru@CoFe-LDH(3%) possesses a lower d-band center that indicates weaker and more optimal binding characteristics for OER intermediates, improving the overall OER performance. Overall, this report displays an excellent correlation between the decorated concentration of NPs over the LDH surface which can tune the OER activity as verified by both experimental and theoretical calculations.

High-content image generation for drug discovery using generative adversarial networks.

Immense amount of high-content image data generated in drug discovery screening requires computationally driven automated analysis. Emergence of advanced machine learning algorithms, like deep learning models, has transformed the interpretation and analysis of imaging data. However, deep learning methods generally require large number of high-quality data samples, which could be limited during preclinical investigations. To address this issue, we propose a generative modeling based computational framework to synthesize images, which can be used for phenotypic profiling of perturbations induced by drug compounds. We investigated the use of three variants of Generative Adversarial Network (GAN) in our framework, viz., a basic Vanilla GAN, Deep Convolutional GAN (DCGAN) and Progressive GAN (ProGAN), and found DCGAN to be most efficient in generating realistic synthetic images. A pre-trained convolutional neural network (CNN) was used to extract features of both real and synthetic images, followed by a classification model trained on real and synthetic images. The quality of synthesized images was evaluated by comparing their feature distributions with that of real images. The DCGAN-based framework was applied to high-content image data from a drug screen to synthesize high-quality cellular images, which were used to augment the real image data. The augmented dataset was shown to yield better classification performance compared with that obtained using only real images. We also demonstrated the application of proposed method on the generation of bacterial images and computed feature distributions for bacterial images specific to different drug treatments. In summary, our results showed that the proposed DCGAN-based framework can be utilized to generate realistic synthetic high-content images, thus enabling the study of drug-induced effects on cells and bacteria.

An Overview on Chemotherapy-induced Cognitive Impairment and Potential Role of Antidepressants.

BACKGROUND: Cognitive impairment is an adverse reaction of cancer chemotherapy and is likely to affect up to 75% of patients during the treatment and 35% of patients experience it for several months after the chemotherapy. Patients manifest symptoms like alteration in working ability, awareness, concentration, visual-verbal memory, attention, executive functions, processing speed, fatigue and behavioural dysfunctions. Post-chemotherapy, cancer survivors have a reduced quality of life due to the symptoms of chemobrain. Apart from this, there are clinical reports which also associate mood disorders, vascular complications, and seizures in some cases. Therefore, the quality of lifestyle of cancer patients/ survivors is severely affected and only worsens due to the absence of any efficacious treatments. With the increase in survivorship, it's vital to identify effective strategies, until then only symptomatic relief for chemobrain can be provided. The depressive symptoms were causally linked to the pathophysiological imbalance between the pro and antiinflammatory cytokines. CONCLUSION: The common causative factor, cytokines can be targeted for the amelioration of an associated symptom of both depression and chemotherapy. Thus, antidepressants can have a beneficial effect on chemotherapy-induced inflammation and cognitive dysfunction via cytokine balance. Also, neurogenesis property of certain antidepressant drugs rationalises their evaluation against CICI. This review briefly glances upon chemotherapy-induced cognitive impairment (CICI), and the modulatory effect of antidepressants on CICI pathomechanisms.