We introduce a comprehensive data-driven framework aimed at enhancing the modeling of physical systems, employing inference techniques and machine-learning enhancements. As a demonstrative application, we pursue the modeling of cathodic electrophoretic deposition, commonly known as e-coating. Our approach illustrates a systematic procedure for enhancing physical models by identifying their limitations through inference on experimental data and introducing adaptable model enhancements to address these shortcomings. We begin by tackling the issue of model parameter identifiability, which reveals aspects of the model that require improvement. To address generalizability, we introduce modifications, which also enhance identifiability. However, these modifications do not fully capture essential experimental behaviors. To overcome this limitation, we incorporate interpretable yet flexible augmentations into the baseline model. These augmentations are parameterized by simple fully-connected neural networks, and we leverage machine-learning tools, particularly neural ordinary differential equations, to learn these augmentations. Our simulations demonstrate that the machine-learning-augmented model more accurately captures observed behaviors and improves predictive accuracy. Nevertheless, we contend that while the model updates offer superior performance and capture the relevant physics, we can reduce off-line computational costs by eliminating certain dynamics without compromising accuracy or interpretability in downstream predictions of quantities of interest, particularly film thickness predictions. The entire process outlined here provides a structured approach to leverage data-driven methods by helping us comprehend the root causes of model inaccuracies and by offering a principled method for enhancing model performance.

The high-power narrow-linewidth fiber laser has become the most widely used high-power laser source nowadays. Further breakthroughs of the output power depend on comprehensive optimization of stimulated Brillouin scattering (SBS), stimulated Raman scattering (SRS) and transverse mode instability (TMI). In this work, we aim to further surpass the power record of all-fiberized and narrow-linewidth fiber amplifiers with near-diffraction-limited (NDL) beam quality. SBS is suppressed by white-noise-signal modulation of a single-frequency seed. In particular, the refractive index of the large-mode-area active fiber in the main amplifier is controlled and fabricated, which could simultaneously increase the effective mode field area of the fundamental mode and the loss coefficient of higher-order modes for balancing SRS and TMI. Subsequent experimental measurements demonstrate a 7.03 kW narrow-linewidth fiber laser with a signal-to-noise ratio of 31.4 dB and beam quality factors of Mx2 = 1.26, My2 = 1.25. To the best of our knowledge, this is the highest reported power with NDL beam quality based on a directly laser-diode-pumped and all-fiberized format, especially with narrow-linewidth spectral emission.

The liver has multiple functions such as detoxification, metabolism, synthesis and storage. Folate is a water-soluble vitamin B9, which participates in one-carbon transfer reactions, maintains methylation capacity and improves oxidative stress. Folic acid is a synthetic form commonly used as a dietary supplement. The liver is the main organ for storing and metabolising folate/folic acid, and the role of folate/folic acid in liver diseases has been widely studied. Deficiency of folate results in methylation capacity dysfunction and can induce liver disorders. However, adverse effects of excessive use of folic acid on the liver have also been reported. This review aims to explore the mechanism of folate/folic acid in different liver diseases, promote further research on folate/folic acid and contribute to its rational clinical application.

Machine learning methods have been used in identifying omics markers for a variety of phenotypes. We aimed to examine whether a supervised machine learning algorithm can improve identification of alcohol-associated transcriptomic markers. In this study, we analysed array-based, whole-blood derived expression data for 17 873 gene transcripts in 5508 Framingham Heart Study participants. By using the Boruta algorithm, a supervised random forest (RF)-based feature selection method, we selected twenty-five alcohol-associated transcripts. In a testing set (30 % of entire study participants), AUC (area under the receiver operating characteristics curve) of these twenty-five transcripts were 0·73, 0·69 and 0·66 for non-drinkers v. moderate drinkers, non-drinkers v. heavy drinkers and moderate drinkers v. heavy drinkers, respectively. The AUC of the selected transcripts by the Boruta method were comparable to those identified using conventional linear regression models, for example, AUC of 1958 transcripts identified by conventional linear regression models (false discovery rate < 0·2) were 0·74, 0·66 and 0·65, respectively. With Bonferroni correction for the twenty-five Boruta method-selected transcripts and three CVD risk factors (i.e. at P < 6·7e-4), we observed thirteen transcripts were associated with obesity, three transcripts with type 2 diabetes and one transcript with hypertension. For example, we observed that alcohol consumption was inversely associated with the expression of DOCK4, IL4R, and SORT1, and DOCK4 and SORT1 were positively associated with obesity, and IL4R was inversely associated with hypertension. In conclusion, using a supervised machine learning method, the RF-based Boruta algorithm, we identified novel alcohol-associated gene transcripts.

Melting and calving of glaciers and ice caps in Antarctica and Greenland could potentially contribute significantly to global sea level rise. Updates to existing outlines that provide critical glacier baseline information in both regions could help in the analysis of particular changes in glacier parameters such as area and volume from time-series inventories. Here we synthesize previously established techniques and apply new multi-source datasets to update glacier outlines in selected test areas of Antarctica and Greenland, as well as to reduce uncertainties and errors during the mapping process. The workflow includes mapping glacier boundaries, subdividing glaciers by watersheds and assigning glacier attributes. Complicated glacier scenarios and updating challenges in polar regions are discussed and demonstrated by representative case studies. For the first time in Antarctica, we analyze the effect of terminus types on mapped glacier areas, and in Greenland we compare the differences with glacier mapping results using Landsat OLI and ETM+. With new data sources, the methods described in this study might help to create glacier outlines on a larger scale in Antarctica and Greenland. Although data sources can be substituted, the enormous amount of manual labor required to update glacier inventories remains a significant challenge.

We investigate experimentally and theoretically the interactions between a cavitation bubble and a hemispherical pendant oil droplet immersed in water. In experiments, the cavitation bubble is generated by a focused laser pulse right below the pendant droplet with well-controlled bubble–wall distances and bubble–droplet size ratios. By high-speed imaging, four typical interactions are observed, namely: oil droplet rupture; water droplet entrapment; oil droplet large deformation; and oil droplet mild deformation. The bubble jetting at the end of collapse and the migration of the bubble centroid are particularly different in each bubble–droplet interaction. We propose theoretical models based on the method of images for calculating the Kelvin impulse and the anisotropy parameter which quantitatively reflects the migration of the bubble centroid at the end of the collapse. Finally, we explain that a combination of the Weber number and the anisotropy parameter determines the regimes of the bubble–droplet interactions.

In this paper, a radio frequency sensor for measuring microfluidics dielectric properties is designed based on microstrip meander line. The meander sensor replaces the straight transmission lines with meander transmission lines in the part of the half-wavelength path difference to improve the sensitivity of the sensor and reduce its size. According to the experimental results, the meander sensor based on the meander line has higher accuracy and a lower relative error than the straight sensor in measuring methanol–ethanol mixtures with different molar fractions. The relative error measured by the meander sensor after calibration with an adjustable cavity is less than 1%. It is easier to detect the very slight variation in dielectric properties brought about by microfluidics. The detection technique can be further applied for the accurate detection of dielectric properties of valuable biological samples, providing a more concise and convenient way.

The sequential occurrence of three layers of smooth muscle layers (SML) in human embryos and fetus is not known. Here, we investigated the process of gut SML development in human embryos and fetuses and compared the morphology of SML in fetuses and neonates. The H&E, Masson trichrome staining, and Immunohistochemistry were conducted on 6–12 gestation week human embryos and fetuses and on normal neonatal intestine. We showed that no lumen was seen in 6–7th gestation week embryonic gut, neither gut wall nor SML was developed in this period. In 8–9th gestation week embryonic and fetal gut, primitive inner circular SML (IC-SML) was identified in a narrow and discontinuous gut lumen with some vacuoles. In 10th gestation week fetal gut, the outer longitudinal SML (OL-SML) in gut wall was clearly identifiable, both the inner and outer SML expressed α-SMA. In 11–12th gestation week fetal gut, in addition to the IC-SML and OL-SML, the muscularis mucosae started to develop as revealed by α-SMA immune-reactivity beneath the developing mucosal epithelial layer. Comparing with the gut of fetuses of 11–12th week of gestation, the muscularis mucosae, IC-SML, and OL-SML of neonatal intestine displayed different morphology, including branching into glands of lamina propria in mucosa and increased thickness. In conclusions, in the human developing gut between week-8 to week-12 of gestation, the IC-SML develops and forms at week-8, followed by the formation of OL-SML at week-10, and the muscularis mucosae develops and forms last at week-12.

Numerous technologies have contributed to the recent development of agriculture, especially the advancement in hyperspectral remote sensing (HRS) constituted a revolution in crop monitoring. The widespread use of HRS to obtain crop parameters suggests the need for a review of research advances in this area. HRS offers new theories and methods for studying crop parameters, but much work needs to be done both experimentally and theoretically before we can truly understand the physical and chemical processes that predict these crop parameters. The study focuses on the following elements: 1) The article provides a relatively comprehensive introduction to HRS and how it can be applied to crop monitoring; 2) Current state-of-the-art techniques are summarized and analyzed to inform further advances in crop monitoring; 3) Opportunities and challenges for crop monitoring applications using HRS are discussed, and future research is summarized. Finally, through a comprehensive discussion and analysis, the article proposes new directions for using HRS to study crop characteristics, such as new data mining techniques including deep learning provide opportunities for efficient processing of large amounts of HRS data; combining the temporal and dynamic characteristics of crop parameters and vegetation growth processes will greatly improve the accuracy of crop parameter detection and monitoring; multidata fusion and multiscale data assimilation will become HRS monitoring. Multidata fusion and multiscale data assimilation will become another research hotspot for HRS monitoring of crop parameters.

The selection of high-quality sperms is critical to intracytoplasmic sperm injection, which accounts for 70–80% of in vitro fertilization (IVF) treatments. So far, sperm screening is usually performed manually by clinicians. However, the performance of manual screening is limited in its objectivity, consistency, and efficiency. To overcome these limitations, we have developed a fast and noninvasive three-stage method to characterize morphology of freely swimming human sperms in bright-field microscopy images using deep learning models. Specifically, we use an object detection model to identify sperm heads, a classification model to select in-focus images, and a segmentation model to extract geometry of sperm heads and vacuoles. The models achieve an F1-score of 0.951 in sperm head detection, a z-position estimation error within ±1.5 μm in in-focus image selection, and a Dice score of 0.948 in sperm head segmentation, respectively. Customized lightweight architectures are used for the models to achieve real-time analysis of 200 frames per second. Comprehensive morphological parameters are calculated from sperm head geometry extracted by image segmentation. Overall, our method provides a reliable and efficient tool to assist clinicians in selecting high-quality sperms for successful IVF. It also demonstrates the effectiveness of deep learning in real-time analysis of live bright-field microscopy images.

The horse played a crucial role in China through the first millennium BC, used both for military advantage and, through incorporation into elite burials, to express social status. Details of how horses were integrated into mortuary contexts during the Qin Empire, however, are poorly understood. Here, the authors present new zooarchaeological data for 24 horses from an accessory pit in Qin Shihuang's mausoleum, indicating that the horses chosen were tall, adult males. These findings provide insights into the selection criteria for animals to be included in the emperor's tomb and invite consideration of questions concerning horse breeds, husbandry practices, and the military and symbolic importance of horses in early imperial China.

We aim to examine the relation of several folate forms (5-methyltetrahydrofolate (5-mTHF), unmetabolised folic acid (UMFA) and MeFox) with kidney function and albuminuria, which remained uncertain. The cross-sectional study was conducted in 18 757 participants from National Health and Nutrition Examination Survey 2011–2018. The kidney outcomes were reduced estimated glomerular filtration rate (eGFR) (<60 ml/min/1·73 m2), microalbuminuria (albumin:creatinine ratio (ACR) of 30–299 mg/g) and macroalbuminuria (ACR ≥ 300 mg/g). Overall, there were significant inverse associations between serum 5-mTHF and kidney outcomes with significant lower prevalence of reduced eGFR (OR, 0·71; 95 % CI: 0·57, 0·87) and macroalbuminuria (OR, 0·65; 95 % CI: 0·46, 0·91) in participants in quartiles 3–4 (v. quartiles 1–2; both Pfor trend across quartiles <0·05). In contrast, there were significant positive relationship between serum UMFA and kidney outcomes with significant higher prevalence of reduced eGFR in participants in quartiles 2–4 (v. quartile 1; OR, 2·12; 95 % CI: 1·45, 3·12; Pfor trend <0·001) and higher prevalence of macroalbuminuria in participants in quartile 4 (v. quartiles 1–3; OR, 1·46; 95 % CI: 1·06, 2·01; Pfor trend <0·001). However, there was no significant associations of 5-mTHF and UMFA with microalbuminuria. In addition, there were significant positive relationships of serum MeFox with reduced eGFR, microalbuminuria and macroalbuminuria (all Pfor trend <0·01). In conclusion, higher 5-mTHF level, along with lower UMFA and MeFox level, was associated with lower prevalence of kidney outcomes, which may help counsel future clinical trials and nutritional guidelines regarding the folate supplement.

Manganese (Mn) oxides have been prevalent on Earth since before the Great Oxidation Event and the Mn cycle is one of the most important biogeochemical processes on the Earth's surface. In sunlit natural environments, the photochemistry of Mn oxides has been discovered to enable solar energy harvesting and conversion in both geological and biological systems. One of the most widespread Mn oxides is birnessite, which is a semiconducting layered mineral that actively drives Mn photochemical cycling in Nature. The oxygen-evolving centre in biological photosystem II (PSII) is also a Mn-cluster of Mn4CaO5, which transforms into a birnessite-like structure during the photocatalytic oxygen evolution process. This phenomenon draws the potential parallel of Mn-functioned photoreactions between the organic and inorganic world. The Mn photoredox cycling involves both the photo-oxidation of Mn(II) and the photoreductive dissolution of Mn(IV/III) oxides. In Nature, the occurrence of Mn(IV/III) photoreduction is usually accompanied with the oxidative degradation of natural organics. For Mn(II) oxidation into Mn oxides, mechanisms of biological catalysis mediated by microorganisms (such as Pseudomonas putida and Bacillus species) and abiotic photoreactions by semiconducting minerals or reactive oxygen species have both been proposed. In particular, anaerobic Mn(II) photo-oxidation processes have been demonstrated experimentally, which shed light on Mn oxide emergence before atmospheric oxygenation on Earth. This review provides a comprehensive and up-to-date elaboration of Mn oxide photoredox cycling in Nature, and gives brand-new insight into the photochemical properties of semiconducting Mn oxides widespread on the Earth's surface.