Morpho-Physiochemical Indices and Transcriptome Analysis Reveal the Role of Glucosinolate and Erucic Acid in Response to Drought Stress during Seed Germination of Rapeseed.

The global expansion of rapeseed seed quality has been focused on maintaining glucosinolate (GSL) and erucic acid (EA) contents. However, the influence of seed GSL and EA contents on the germination process under drought stress remains poorly understood. Herein, 114 rapeseed accessions were divided into four groups based on GSL and EA contents to investigate their performance during seed imbibition under drought stress. Our results revealed significant variations in seed germination-related traits, particularly with higher GSL and EA, which exhibited higher germination % (G%) and lower mean germination time (MGT) under drought stress conditions. Moreover, osmoregulation, enzymatic system and hormonal regulation were improved in high GSL and high EA (HGHE) versus low GSL and low EA (LGLE) seeds, indicating the essential protective role of GSL and EA during the germination process in response to drought stress. The transcriptional regulation mechanism for coordinating GSL-EA-related pathways in response to drought stress during seed imbibition was found to involve the differential expression of sugar metabolism-, antioxidant-, and hormone-related genes with higher enrichment in HGHE compared to LGLE seeds. GO enrichment analysis showed higher variations in transcription regulator activity and DNA-binding transcription factors, as well as ATP and microtubule motor activity in GSL-EA-related pathways. Furthermore, KEGG analysis identified cellular processes, environmental information processing, and metabolism categories, with varied gene participation between GSL, EA and GSL-EA-related pathways. For further clarification, QY7 (LGLE) seeds were primed with different concentrations of GSL and EA under drought stress conditions. The results showed that 200 µmol/L of GSL and 400 µmol/L of EA significantly improved G%, MGT, and seedling fresh weight, besides regulating stress and fatty acid responsive genes during the seed germination process under drought stress conditions. Conclusively, exogenous application of GSL and EA is considered a promising method for enhancing the drought tolerance of LGLE seeds. Furthermore, the current investigation could provide a theoretical basis of GSL and EA roles and their underlying mechanisms in stress tolerance during the germination process.

Phytoremediation of copper-contaminated soils by rapeseed (Brassica napus L.) and underlying molecular mechanisms for copper absorption and sequestration.

High levels of copper released in the soil, mainly from anthropogenic activity, can be hazardous to plants, animals, and humans. The present research aimed to estimate the suitability and effectiveness of rapeseed (Brassica napus L.) as a possible soil remediation option and to uncover underlying adaptive mechanisms A pot experiment was conducted to explore the effect of copper stress on agronomic and yield traits for 32 rapeseed genotypes. The copper-tolerant genotype H2009 and copper-sensitive genotype ZYZ16 were selected for further physiological, metabolomic, and transcriptomic analyses. The results exhibited a significant genotypic variation in copper stress tolerance in rapeseed. Specifically, the ratio of seed yield under copper stress to control ranged from 0.29 to 0.74. Furthermore, the proline content and antioxidant enzymatic activities in the roots were greater than those in the shoots. The accumulated copper in the roots accounted for about 50% of the total amount absorbed by plants; thus, the genotypes possessing high root volumes can be used for rhizofiltration to uptake and sequester copper. Additionally, the pectin and hemicellulose contents were significantly increased by 15.6% and 162%, respectively, under copper stress for the copper-tolerant genotype, allowing for greater sequestration of copper ions in the cell wall and lower oxidative stress. Comparative analysis of transcriptomes and metabolomes revealed that excessive copper enhanced the up-regulation of functional genes or metabolites related to cell wall binding, copper transportation, and chelation in the copper-tolerant genotype. Our results suggest that copper-tolerant rapeseed can thrive in heavily copper-polluted soils with a 5.85% remediation efficiency as well as produce seed and vegetable oil without exceeding food quality standards for the industry. This multi-omics comparison study provides insights into breeding copper-tolerant genotypes that can be used for the phytoremediation of heavy metal-polluted soils.

Inferring neural activity before plasticity as a foundation for learning beyond backpropagation.

For both humans and machines, the essence of learning is to pinpoint which components in its information processing pipeline are responsible for an error in its output, a challenge that is known as 'credit assignment'. It has long been assumed that credit assignment is best solved by backpropagation, which is also the foundation of modern machine learning. Here, we set out a fundamentally different principle on credit assignment called 'prospective configuration'. In prospective configuration, the network first infers the pattern of neural activity that should result from learning, and then the synaptic weights are modified to consolidate the change in neural activity. We demonstrate that this distinct mechanism, in contrast to backpropagation, (1) underlies learning in a well-established family of models of cortical circuits, (2) enables learning that is more efficient and effective in many contexts faced by biological organisms and (3) reproduces surprising patterns of neural activity and behavior observed in diverse human and rat learning experiments.

Cross-domain attention-guided generative data augmentation for medical image analysis with limited data.

Data augmentation is widely applied to medical image analysis tasks in limited datasets with imbalanced classes and insufficient annotations. However, traditional augmentation techniques cannot supply extra information, making the performance of diagnosis unsatisfactory. GAN-based generative methods have thus been proposed to obtain additional useful information to realize more effective data augmentation; but existing generative data augmentation techniques mainly encounter two problems: (i) Current generative data augmentation lacks of the capability in using cross-domain differential information to extend limited datasets. (ii) The existing generative methods cannot provide effective supervised information in medical image segmentation tasks. To solve these problems, we propose an attention-guided cross-domain tumor image generation model (CDA-GAN) with an information enhancement strategy. The CDA-GAN can generate diverse samples to expand the scale of datasets, improving the performance of medical image diagnosis and treatment tasks. In particular, we incorporate channel attention into a CycleGAN-based cross-domain generation network that captures inter-domain information and generates positive or negative samples of brain tumors. In addition, we propose a semi-supervised spatial attention strategy to guide spatial information of features at the pixel level in tumor generation. Furthermore, we add spectral normalization to prevent the discriminator from mode collapse and stabilize the training procedure. Finally, to resolve an inapplicability problem in the segmentation task, we further propose an application strategy of using this data augmentation model to achieve more accurate medical image segmentation with limited data. Experimental studies on two public brain tumor datasets (BraTS and TCIA) show that the proposed CDA-GAN model greatly outperforms the state-of-the-art generative data augmentation in both practical medical image classification tasks and segmentation tasks; e.g. CDA-GAN is 0.50%, 1.72%, 2.05%, and 0.21% better than the best SOTA baseline in terms of ACC, AUC, Recall, and F1, respectively, in the classification task of BraTS, while its improvements w.r.t. the best SOTA baseline in terms of Dice, Sens, HD95, and mIOU, in the segmentation task of TCIA are 2.50%, 0.90%, 14.96%, and 4.18%, respectively.

Multi-ConDoS: Multimodal Contrastive Domain Sharing Generative Adversarial Networks for Self-Supervised Medical Image Segmentation.

Existing self-supervised medical image segmentation usually encounters the domain shift problem (i.e., the input distribution of pre-training is different from that of fine-tuning) and/or the multimodality problem (i.e., it is based on single-modal data only and cannot utilize the fruitful multimodal information of medical images). To solve these problems, in this work, we propose multimodal contrastive domain sharing (Multi-ConDoS) generative adversarial networks to achieve effective multimodal contrastive self-supervised medical image segmentation. Compared to the existing self-supervised approaches, Multi-ConDoS has the following three advantages: (i) it utilizes multimodal medical images to learn more comprehensive object features via multimodal contrastive learning; (ii) domain translation is achieved by integrating the cyclic learning strategy of CycleGAN and the cross-domain translation loss of Pix2Pix; (iii) novel domain sharing layers are introduced to learn not only domain-specific but also domain-sharing information from the multimodal medical images. Extensive experiments on two publicly multimodal medical image segmentation datasets show that, with only 5% (resp., 10%) of labeled data, Multi-ConDoS not only greatly outperforms the state-of-the-art self-supervised and semi-supervised medical image segmentation baselines with the same ratio of labeled data, but also achieves similar (sometimes even better) performances as fully supervised segmentation methods with 50% (resp., 100%) of labeled data, which thus proves that our work can achieve superior segmentation performances with very low labeling workload. Furthermore, ablation studies prove that the above three improvements are all effective and essential for Multi-ConDoS to achieve this very superior performance.

Automatic data augmentation for medical image segmentation using Adaptive Sequence-length based Deep Reinforcement Learning.

Although existing deep reinforcement learning-based approaches have achieved some success in image augmentation tasks, their effectiveness and adequacy for data augmentation in intelligent medical image analysis are still unsatisfactory. Therefore, we propose a novel Adaptive Sequence-length based Deep Reinforcement Learning (ASDRL) model for Automatic Data Augmentation (AutoAug) in intelligent medical image analysis. The improvements of ASDRL-AutoAug are two-fold: (i) To remedy the problem of some augmented images being invalid, we construct a more accurate reward function based on different variations of the augmentation trajectories. This reward function assesses the validity of each augmentation transformation more accurately by introducing different information about the validity of the augmented images. (ii) Then, to alleviate the problem of insufficient augmentation, we further propose a more intelligent automatic stopping mechanism (ASM). ASM feeds a stop signal to the agent automatically by judging the adequacy of image augmentation. This ensures that each transformation before stopping the augmentation can smoothly improve the model performance. Extensive experimental results on three medical image segmentation datasets show that (i) ASDRL-AutoAug greatly outperforms the state-of-the-art data augmentation methods in medical image segmentation tasks, (ii) the proposed improvements are both effective and essential for ASDRL-AutoAug to achieve superior performance, and the new reward evaluates the transformations more accurately than existing reward functions, and (iii) we also demonstrate that ASDRL-AutoAug is adaptive for different images in terms of sequence length, as well as generalizable across different segmentation models.

Hybrid Reinforced Medical Report Generation With M-Linear Attention and Repetition Penalty.

To reduce doctors' workload, deep-learning-based automatic medical report generation has recently attracted more and more research efforts, where deep convolutional neural networks (CNNs) are employed to encode the input images, and recurrent neural networks (RNNs) are used to decode the visual features into medical reports automatically. However, these state-of-the-art methods mainly suffer from three shortcomings: 1) incomprehensive optimization; 2) low-order and unidimensional attention; and 3) repeated generation. In this article, we propose a hybrid reinforced medical report generation method with m-linear attention and repetition penalty mechanism (HReMRG-MR) to overcome these problems. Specifically, a hybrid reward with different weights is employed to remedy the limitations of single-metric-based rewards, and a local optimal weight search algorithm is proposed to significantly reduce the complexity of searching the weights of the rewards from exponential to linear. Furthermore, we use m-linear attention modules to learn multidimensional high-order feature interactions and to achieve multimodal reasoning, while a new repetition penalty is proposed to apply penalties to repeated terms adaptively during the model's training process. Extensive experimental studies on two public benchmark datasets show that HReMRG-MR greatly outperforms the state-of-the-art baselines in terms of all metrics. The effectiveness and necessity of all components in HReMRG-MR are also proved by ablation studies. Additional experiments are further conducted and the results demonstrate that our proposed local optimal weight search algorithm can significantly reduce the search time while maintaining superior medical report generation performances.

Nitrogen supply alleviates seed yield reduction by improving the morphology and carbon metabolism of pod walls in shaded rapeseed.

Shading significantly affects rapeseed yield, while reasonable nitrogen (N) application has efficiency gains. However, the functions and mechanisms of N are not fully established for shaded rapeseed plants. Therefore, we conducted a 2-year field experiment to study the effect of N on pod wall morphology and carbon metabolism of shaded rapeseed. Two varieties, three N rates (120 [N1], 240 [N2], and 360 [N3] kg hm-2 ) and two light intensities (100 and 70% light transmission) from 10 to 35 days after the end of flowering were set as experimental parameters. Shading decreased the pod wall chlorophyll content, ribulose 1,5-bisphosphate carboxylase (Rubisco) activity and glucose content at 25 and 35 days after flowering (DAF). Decreased sucrose synthase (SuSy) and sucrose phosphate synthase activity caused by shading reduced sucrose and fructose content. They are responsible for the decline in the 1000-seed weight and a 22.1-37.6% decline in seed yield. More N under shading promoted pod elongation and pigment content, improved chloroplast ultrastructure, increased Rubisco and SuSy activity at 35 DAF, thus contributing to pod wall photosynthesis and fructose and glucose levels in shaded rapeseed plants. Similar trends were observed in pod number, pod weight, and seed weight, while the greatest increase in seed/wall ratio was observed under N2 for shaded rapeseed plants. The results indicated that N can reduce the yield difference between different light conditions and balance partitioning and conversion of photoassimilates in pod wall, but avoid applying an excessive amount of nitrogen.

17ß-estradiol inhibits TGF-ß-induced collagen gel contraction mediated by human Tenon fibroblasts via Smads and MAPK signaling pathways.

AIM: To investigate the impact of 17ß-estradiol on the collagen gels contraction (CGC) and inflammation induced by transforming growth factor (TGF)-ß in human Tenon fibroblasts (HTFs). METHODS: HTFs were three-dimensionally cultivated in type I collagen-generated gels with or without TGF-ß (5 ng/mL), 17ß-estradiol (12.5 to 100 µmol/L), or progesterone (12.5 to 100 µmol/L). Then, the collagen gel diameter was determined to assess the contraction, and the development of stress fibers was analyzed using immunofluorescence staining. Immunoblot and gelatin zymography assays were used to analyze matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) being released into culture supernatants. Enzyme-linked immunosorbent assay (ELISA) and reverse transcription-quantitative polymerase chain reaction (RT-PCR) were used to detect interleukin (IL)-6, monocyte chemoattractant proteins (MCP)-1, and vascular endothelial growth factor (VEGF) in HTFs at the translational and transcriptional levels. The phosphorylation levels of Sma- and Mad-related proteins (Smads), mitogen-activated protein kinases (MAPKs), and protein kinase B (AKT) were measured by immunoblotting. Statistical analysis was performed using either the Tukey-Kramer test or Student's unpaired t-test to compare the various treatments. RESULTS: The CGC caused by TGF-ß in HTFs was significantly inhibited by 17ß-estradiol (25 to 100 µmol/L), and a statistically significant difference was observed when comparing the normal control group with 17ß-estradiol concentrations exceeding 25 µmol/L (P<0.05). The suppressive impact of 17ß-estradiol became evident 24h after administration and peaked at 72h (P<0.05), whereas progesterone had no impact. Moreover, 17ß-estradiol attenuated the formation of stress fibers, and the production of MMP-3 and MMP-1 in HTFs stimulated by TGF-ß. The expression of MCP-1, IL-6, and VEGF mRNA and protein in HTFs were suppressed by 100 µmol/L 17ß-estradiol (P<0.01). Additionally, the phosphorylation of Smad2 Smad3, p38, and extracellular signal-regulated kinase (ERK) were downregulated (P <0.01). CONCLUSION: 17ß-estradiol significantly inhibits the CGC and inflammation caused by TGF-ß in HTFs. This inhibition is likely related to the suppression of stress fibers, inhibition of MMPs, and attenuation of Smads and MAPK (ERK and p38) signaling. 17ß-estradiol may have potential clinical benefits in preventing scar development and inflammation in the conjunctiva.

Protection of Barrier Function in Cultured Human Corneal Epithelial Cells by Semaphorin 4D.

PURPOSE: Corneal epithelial barrier function is important to maintain corneal homeostasis and is impaired by inflammation. We aimed to investigate the localization of semaphorin 4D (Sema4D) in the cornea, and its effects on the barrier function of cultured corneal epithelial cells. METHODS: The expressions of semaphorin4 D and its receptor in the murine cornea were examined by immunoblot, immunofluorescent staining and confocal microscopy observations. Human corneal epithelial (HCE) cells stimulated by TNF-α or IL-1ß were cultured with or without Sema4D. Cell viability was examined by CCK8, cell migration was evaluated by scratch wound assay, and barrier function was assessed by transepithelial electrical resistance (TEER) and Dextran-FITC permeability assay. The expression of tight junction proteins in HCE cells was examined by immunoblot, immunofluorescent staining and qRT-PCR. RESULTS: We demonstrated that the protein of Sema4D and its receptor, plexin-B1, was expressed in murine cornea. Sema4D induced an increase in the TEER and a decrease in the permeability of HCE cells. It also induced the expression of tight junction protein ZO-1, occludin and claudin-1 in HCE cells. Furthermore, under stimulation of TNF-α or IL-1ß, Sema4D treatment could inhibit the decreased TEER and the elevated permeability of HCE cells. CONCLUSIONS: Sema4D is located distinctly in corneal epithelial cells and promoted their barrier function by increasing the expression of tight junction proteins. Sema4D may act as a preventive for maintaining corneal epithelial barrier function during ocular inflammation.

EFPN: Effective medical image detection using feature pyramid fusion enhancement.

Feature pyramid networks (FPNs) are widely used in the existing deep detection models to help them utilize multi-scale features. However, there exist two multi-scale feature fusion problems for the FPN-based deep detection models in medical image detection tasks: insufficient multi-scale feature fusion and the same importance for multi-scale features. Therefore, in this work, we propose a new enhanced backbone model, EFPNs, to overcome these problems and help the existing FPN-based detection models to achieve much better medical image detection performances. We first introduce an additional top-down pyramid to help the detection networks fuse deeper multi-scale information; then, a scale enhancement module is developed to use different sizes of kernels to generate more diverse multi-scale features. Finally, we propose a feature fusion attention module to estimate and assign different importance weights to features with different depths and scales. Extensive experiments are conducted on two public lesion detection datasets for different medical image modalities (X-ray and MRI). On the mAP and mR evaluation metrics, EFPN-based Faster R-CNNs improved 1.55% and 4.3% on the PenD (X-ray) dataset, and 2.74% and 3.1% on the BraTs (MRI) dataset, respectively. EFPN-based Faster R-CNNs achieve much better performances than the state-of-the-art baselines in medical image detection tasks. The proposed three improvements are all essential and effective for EFPNs to achieve superior performances; and besides Faster R-CNNs, EFPNs can be easily applied to other deep models to significantly enhance their performances in medical image detection tasks.

Accurate and intelligent diagnosis of pediatric pneumonia using X-ray images and blood testing data.

Computer-aided diagnosis (CAD) methods such as the X-rays-based method is one of the cheapest and safe alternative options to diagnose the disease compared to other alternatives such as Computed Tomography (CT) scan, and so on. However, according to our experiments on X-ray public datasets and real clinical datasets, we found that there are two challenges in the current classification of pneumonia: existing public datasets have been preprocessed too well, making the accuracy of the results relatively high; existing models have weak ability to extract features from the clinical pneumonia X-ray dataset. To solve the dataset problems, we collected a new dataset of pediatric pneumonia with labels obtained through a comprehensive pathogen-radiology-clinical diagnostic screening. Then, to accurately capture the important features in imbalanced data, based on the new dataset, we proposed for the first time a two-stage training multimodal pneumonia classification method combining X-ray images and blood testing data, which improves the image feature extraction ability through a global-local attention module and mitigate the influence of class imbalance data on the results through the two-stage training strategy. In experiments, the performance of our proposed model is the best on new clinical data and outperforms the diagnostic accuracy of four experienced radiologists. Through further research on the performance of various blood testing indicators in the model, we analyzed the conclusions that are helpful for radiologists to diagnose.

µ-Net: Medical image segmentation using efficient and effective deep supervision.

Although the existing deep supervised solutions have achieved some great successes in medical image segmentation, they have the following shortcomings; (i) semantic difference problem: since they are obtained by very different convolution or deconvolution processes, the intermediate masks and predictions in deep supervised baselines usually contain semantics with different depth, which thus hinders the models' learning capabilities; (ii) low learning efficiency problem: additional supervision signals will inevitably make the training of the models more time-consuming. Therefore, in this work, we first propose two deep supervised learning strategies, U-Net-Deep and U-Net-Auto, to overcome the semantic difference problem. Then, to resolve the low learning efficiency problem, upon the above two strategies, we further propose a new deep supervised segmentation model, called µ-Net, to achieve not only effective but also efficient deep supervised medical image segmentation by introducing a tied-weight decoder to generate pseudo-labels with more diverse information and also speed up the convergence in training. Finally, three different types of µ-Net-based deep supervision strategies are explored and a Similarity Principle of Deep Supervision is further derived to guide future research in deep supervised learning. Experimental studies on four public benchmark datasets show that µ-Net greatly outperforms all the state-of-the-art baselines, including the state-of-the-art deeply supervised segmentation models, in terms of both effectiveness and efficiency. Ablation studies sufficiently prove the soundness of the proposed Similarity Principle of Deep Supervision, the necessity and effectiveness of the tied-weight decoder, and using both the segmentation and reconstruction pseudo-labels for deep supervised learning.

PAC-Net: Multi-pathway FPN with position attention guided connections and vertex distance IoU for 3D medical image detection.

Automatic medical image detection aims to utilize artificial intelligence techniques to detect lesions in medical images accurately and efficiently, which is one of the most important tasks in computer-aided diagnosis (CAD) systems, and can be embedded into portable imaging devices for intelligent Point of Care (PoC) Diagnostics. The Feature Pyramid Networks (FPN) based models are widely used deep-learning-based solutions for automatic medical image detection. However, FPN-based medical lesion detection models have two shortcomings: the object position offset problem and the degradation problem of IoU-based loss. Therefore, in this work, we propose a novel FPN-based backbone model, i.e., Multi-Pathway Feature Pyramid Networks with Position Attention Guided Connections and Vertex Distance IoU (abbreviated as PAC-Net), to replace vanilla FPN for more accurate lesion detection, where two innovative improvements, a position attention guided connection (PAC) module and Vertex Distance IoU Vertex Distance Intersection over Union loss, are proposed to address the above-mentioned shortcomings of vanilla FPN, respectively. Extensive experiments are conducted on a public medical image detection dataset, i.e., Deeplesion, and the results showed that i) PAC-Net outperforms all state-of-the-art FPN-based depth models in both evaluation metrics of lesion detection on the DeepLesion dataset, ii) the proposed PAC module and VDIoU loss are both effective and important for PAC-Net to achieve a superior performance in automatic medical image detection tasks, and iii) the proposed VDIoU loss converges more quickly than the existing IoU-based losses, making PAC-Net an accurate and also highly efficient 3D medical image detection model.

Painless and accurate medical image analysis using deep reinforcement learning with task-oriented homogenized automatic pre-processing.

Pre-processing is widely applied in medical image analysis to remove the interference information. However, the existing pre-processing solutions mainly encounter two problems: (i) it is heavily relied on the assistance of clinical experts, making it hard for intelligent CAD systems to deploy quickly; (ii) due to the personnel and information barriers, it is difficult for medical institutions to conduct the same pre-processing operations, making a deep model that performs well on a specific medical institution difficult to achieve similar performances on the same task in other medical institutions. To overcome these problems, we propose a deep-reinforcement-learning-based task-oriented homogenized automatic pre-processing (DRL-HAPre) framework to overcome these two problems. This framework utilizes deep reinforcement learning techniques to learn a policy network to automatically and adaptively select the optimal pre-processing operations for the input medical images according to different analysis tasks, thus helping the intelligent CAD system to achieve a rapid deployment (i.e., painless) and maintain a satisfactory performance (i.e., accurate) among different medical institutes. To verify the effectiveness and advantages of the proposed DRL-HAPre framework, we further develop a homogenized automatic pre-processing model based on the DRL-HAPre framework to realize the automatic pre-processing of key region selection (called HAPre-KRS) in the pneumonia image classification task. Extensive experimental studies are conducted on three pediatric pneumonia classification datasets with different image qualities, and the results show that: (i) There does exist a hard-to-reproduce problem in clinical practices and the fact that having different medical image qualities in different medical institutes is an important reason for the existing of hard-to-reproduce problem, so it is compelling to propose homogenized automatic pre-processing method. (ii) The proposed HAPre-KRS model and DRL-HAPre framework greatly outperform three kinds of state-of-the-art baselines (i.e., pre-processing, attention and pneumonia baseline), and the lower the medical image quality, the greater the improvements of using our HAPre-KRS model and DRL-HAPre framework. (iii) With the help of homogenized pre-processing, HAPre-KRS (and DRL-HAPre framework) can greatly avoid performance degradation in real-world cross-source applications (i.e., thus overcoming the hard-to-reproduce problem).

An ultra-sensitive photothermal lateral flow immunoassay for 17ß-estradiol in food samples.

Hormone residues in food and drinking water endanger human health, therefore, on-site analysis techniques of superior performance are important for monitoring this risk. In this study, an ultra-sensitive photothermal lateral flow immunoassay (LFIA) for quantification of 17ß-estradiol (E2) has been developed. Anti-E2 antibody modified black phosphorus-Au (BP-Au) nanocomposite was developed as a photothermal contrast signal probe and the temperature at test-zone was recorded with an infrared camera. Under the irradiation of 808 nm laser at test-zone, it gave temperatures negatively related to the concentrations of E2 in samples. Under optimal detecting conditions, the developed photothermal LFIA exhibited a limit of detection of 50 pg mL-1, over 100-fold more sensitive than visual LFIA, and a linear range of 3 orders of magnitude. This method has been successfully applied to water, milk, and milk powder samples.

Multi-modal contrastive mutual learning and pseudo-label re-learning for semi-supervised medical image segmentation.

Semi-supervised learning has a great potential in medical image segmentation tasks with a few labeled data, but most of them only consider single-modal data. The excellent characteristics of multi-modal data can improve the performance of semi-supervised segmentation for each image modality. However, a shortcoming for most existing multi-modal solutions is that as the corresponding processing models of the multi-modal data are highly coupled, multi-modal data are required not only in the training but also in the inference stages, which thus limits its usage in clinical practice. Consequently, we propose a semi-supervised contrastive mutual learning (Semi-CML) segmentation framework, where a novel area-similarity contrastive (ASC) loss leverages the cross-modal information and prediction consistency between different modalities to conduct contrastive mutual learning. Although Semi-CML can improve the segmentation performance of both modalities simultaneously, there is a performance gap between two modalities, i.e., there exists a modality whose segmentation performance is usually better than that of the other. Therefore, we further develop a soft pseudo-label re-learning (PReL) scheme to remedy this gap. We conducted experiments on two public multi-modal datasets. The results show that Semi-CML with PReL greatly outperforms the state-of-the-art semi-supervised segmentation methods and achieves a similar (and sometimes even better) performance as fully supervised segmentation methods with 100% labeled data, while reducing the cost of data annotation by 90%. We also conducted ablation studies to evaluate the effectiveness of the ASC loss and the PReL module.

The Deterioration of Agronomical Traits of the Continuous Cropping of Stevia Is Associated With the Dynamics of Soil Bacterial Community.

Stevia rebaudiana Bertoni is grown worldwide as an important, natural sweetener resource plant. The yield of steviol glycosides (SVglys) is greatly influenced by continuous cropping. In this study, we collected the roots, rhizosphere soils, and bulk soils from 2 years of continuous cropping (Y2) and 8 years of continuous cropping (Y8). A high-throughput sequencing technology based on Illumina Hiseq 2500 platform was used to study the structure and diversity of bacterial communities in the roots and soils of stevia with different years of continuous cropping. The results demonstrated that although the content of a group of SVglys was significantly increased in stevia of long-term continuous cropping, it inhibited the growth of plants and lowered the leaf dry weight; as a result, the total amount of SVglys was significantly decreased. Meanwhile, continuous cropping changed the physicochemical properties and the bacterial composition communities of soil. The different sampling sources of the root, rhizosphere soil, and bulk soil had no impact on the richness of bacterial communities, while it exhibited obvious effects on the diversity of bacterial communities. Continuous cropping had a stronger effect on the bacterial community composition in rhizosphere soil than in root and bulk soil. Based on linear discriminant analysis effect size (LEfSe), in the rhizosphere soil of Y8, the relative abundance of some beneficial bacterial genera of Sphingomonas, Devosia, Streptomyces, and Flavobacterium decreased significantly, while the relative abundance of Polycyclovorans, Haliangium, and Nitrospira greatly increased. Moreover, the soil pH and nutrient content, especially the soil organic matter, were correlated with the relative abundance of predominant bacteria at the genus level. This study provides a theoretical basis for uncovering the mechanism of obstacles in continuous stevia cropping and provides guidance for the sustainable development of stevia.

Metal-organic framework mixed-matrix membrane-based extraction combined HPLC for determination of bisphenol A in milk and milk packaging.

The residues of bisphenol A (BPA) in milk packaging may transfer to milk, adversely affecting the human endocrine system. Consequently, to analyse or monitor BPA, it is imperative to develop rapid and effective approaches to BPA extraction from milk and milk packing as BPA is usually present in trace levels. Herein, we developed a rapid, simple, and low-cost dispersive-membrane-solid-phase-extraction (DME) for BPA with MIL-101(Cr) mixed-matrix-membrane (MMM). The MMM had large surface area (1322.09 m2/g) and pore volume (0.65 cm3/g), possessed great extraction efficiency of BPA, and kept more than 90% extraction efficiency after 20 times of reuse. Using the developed MIL-101(Cr)-MMM-based DME coupled with HPLC-fluorescence detector, we received an adequate linearity in the range of 0.1 âˆ¼ 50 µg/L BPA and a limit of detection as low as 16 ng/L under optimized conditions. The recoveries of BPA in milk and milk bottles were from 74.2% to 110.6%, with RSDs less than 9.4%.

Reverse Differentiation via Predictive Coding.

Deep learning has redefined AI thanks to the rise of artificial neural networks, which are inspired by neuronal networks in the brain. Through the years, these interactions between AI and neuroscience have brought immense benefits to both fields, allowing neural networks to be used in a plethora of applications. Neural networks use an efficient implementation of reverse differentiation, called backpropagation (BP). This algorithm, however, is often criticized for its biological implausibility (e.g., lack of local update rules for the parameters). Therefore, biologically plausible learning methods that rely on predictive coding (PC), a framework for describing information processing in the brain, are increasingly studied. Recent works prove that these methods can approximate BP up to a certain margin on multilayer perceptrons (MLPs), and asymptotically on any other complex model, and that zerodivergence inference learning (Z-IL), a variant of PC, is able to exactly implement BP on MLPs. However, the recent literature shows also that there is no biologically plausible method yet that can exactly replicate the weight update of BP on complex models. To fill this gap, in this paper, we generalize (PC and) Z-IL by directly defining it on computational graphs, and show that it can perform exact reverse differentiation. What results is the first PC (and so biologically plausible) algorithm that is equivalent to BP in the way of updating parameters on any neural network, providing a bridge between the interdisciplinary research of neuroscience and deep learning. Furthermore, the above results in particular also immediately provide a novel local and parallel implementation of BP.