Multiview Spatial-Temporal Meta-Learning for Multivariate Time Series Forecasting.

Multivariate time series modeling has been essential in sensor-based data mining tasks. However, capturing complex dynamics caused by intra-variable (temporal) and inter-variable (spatial) relationships while simultaneously taking into account evolving data distributions is a non-trivial task, which faces accumulated computational overhead and multiple temporal patterns or distribution modes. Most existing methods focus on the former direction without adaptive task-specific learning ability. To this end, we developed a holistic spatial-temporal meta-learning probabilistic inference framework, entitled ST-MeLaPI, for the efficient and versatile learning of complex dynamics. Specifically, first, a multivariate relationship recognition module is utilized to learn task-specific inter-variable dependencies. Then, a multiview meta-learning and probabilistic inference strategy was designed to learn shared parameters while enabling the fast and flexible learning of task-specific parameters for different batches. At the core are spatial dependency-oriented and temporal pattern-oriented meta-learning approximate probabilistic inference modules, which can quickly adapt to changing environments via stochastic neurons at each timestamp. Finally, a gated aggregation scheme is leveraged to realize appropriate information selection for the generative style prediction. We benchmarked our approach against state-of-the-art methods with real-world data. The experimental results demonstrate the superiority of our approach over the baselines.

GC-STCL: A Granger Causality-Based Spatial-Temporal Contrastive Learning Framework for EEG Emotion Recognition.

EEG signals capture information through multi-channel electrodes and hold promising prospects for human emotion recognition. However, the presence of high levels of noise and the diverse nature of EEG signals pose significant challenges, leading to potential overfitting issues that further complicate the extraction of meaningful information. To address this issue, we propose a Granger causal-based spatial-temporal contrastive learning framework, which significantly enhances the ability to capture EEG signal information by modeling rich spatial-temporal relationships. Specifically, in the spatial dimension, we employ a sampling strategy to select positive sample pairs from individuals watching the same video. Subsequently, a Granger causality test is utilized to enhance graph data and construct potential causality for each channel. Finally, a residual graph convolutional neural network is employed to extract features from EEG signals and compute spatial contrast loss. In the temporal dimension, we first apply a frequency domain noise reduction module for data enhancement on each time series. Then, we introduce the Granger-Former model to capture time domain representation and calculate the time contrast loss. We conduct extensive experiments on two publicly available sentiment recognition datasets (DEAP and SEED), achieving 1.65% improvement of the DEAP dataset and 1.55% improvement of the SEED dataset compared to state-of-the-art unsupervised models. Our method outperforms benchmark methods in terms of prediction accuracy as well as interpretability.

Development of a multispectral fluorescence LiDAR for point cloud segmentation of plants.

The accelerating development of high-throughput plant phenotyping demands a LiDAR system to achieve spectral point cloud, which will significantly improve the accuracy and efficiency of segmentation based on its intrinsic fusion of spectral and spatial data. Meanwhile, a relatively longer detection range is required for platforms e.g., unmanned aerial vehicles (UAV) and poles. Towards the aims above, what we believe to be, a novel multispectral fluorescence LiDAR, featuring compact volume, light weight, and low cost, has been proposed and designed. A 405 nm laser diode was employed to excite the fluorescence of plants, and the point cloud attached with both the elastic and inelastic signal intensities that was obtained through the R-, G-, B-channels of a color image sensor. A new position retrieval method has been developed to evaluate far field echo signals, from which the spectral point cloud can be obtained. Experiments were designed to validate the spectral/spatial accuracy and the segmentation performance. It has been found out that the values obtained through the R-, G-, B-channels are consistent with the emission spectrum measured by a spectrometer, achieving a maximum R2 of 0.97. The theoretical spatial resolution can reach up to 47 mm and 0.7 mm in the x- and y-direction at a distance of around 30 m, respectively. The values of recall, precision, and F score for the segmentation of the fluorescence point cloud were all beyond 0.97. Besides, a field test has been carried out on plants at a distance of about 26 m, which further demonstrated that the multispectral fluorescence data can significantly facilitate the segmentation process in a complex scene. These promising results prove that the proposed multispectral fluorescence LiDAR has great potential in applications of digital forestry inventory and intelligent agriculture.

Design, synthesis, and biological evaluation of THIQ as antidepressive agents.

7-substituted tetrahydroisoquinolines derivatives were designed, synthesized, and evaluated for neuroprotective properties. We summarized the preliminary structure activity relationships (SAR). Compound 3i was screened as a hit compound and its antidepressant activity was evaluated by employing the forced swimming test, tail suspension test. Additionally, ADMET profile (absorption, distribution, metabolism, excretion and toxicity properties) of the compound 3i was predicted in order to evaluate their lead-like properties and safety. The interaction of compound 3i bound to MAO-A was explored using molecular docking and molecular dynamics simulation. Results of biological studies revealed that the compound 3i exhibited almost equal antidepressant activity compared with magnoflorine. Compound 3i is predicted to possess good drug like properties and safety based on ADMET profile predictions. This work provides ideas for the drugs discovery of antidepressant agents.

KAMPNet: multi-source medical knowledge augmented medication prediction network with multi-level graph contrastive learning.

BACKGROUNDS: Predicting medications is a crucial task in intelligent healthcare systems, aiding doctors in making informed decisions based on electronic medical records (EMR). However, medication prediction faces challenges due to complex relations within heterogeneous medical data. Existing studies primarily focus on the supervised mining of hierarchical relations between homogeneous codes in medical ontology graphs, such as diagnosis codes. Few studies consider the valuable relations, including synergistic relations between medications, concurrent relations between diseases, and therapeutic relations between medications and diseases from historical EMR. This limitation restricts prediction performance and application scenarios. METHODS: To address these limitations, we propose KAMPNet, a multi-sourced medical knowledge augmented medication prediction network. KAMPNet captures diverse relations between medical codes using a multi-level graph contrastive learning framework. Firstly, unsupervised graph contrastive learning with a graph attention network encoder captures implicit relations within homogeneous medical codes from the medical ontology graph, generating knowledge augmented medical code embedding vectors. Then, unsupervised graph contrastive learning with a weighted graph convolutional network encoder captures correlative relations between homogeneous or heterogeneous medical codes from the constructed medical codes relation graph, producing relation augmented medical code embedding vectors. Finally, the augmented medical code embedding vectors, along with supervised medical code embedding vectors, are fed into a sequential learning network to capture temporal relations of medical codes and predict medications for patients. RESULTS: Experimental results on the public MIMIC-III dataset demonstrate the superior performance of our KAMPNet model over several baseline models, as measured by Jaccard, F1 score, and PR-AUC for medication prediction. CONCLUSIONS: Our KAMPNet model can effectively capture the valuable relations between medical codes inherent in multi-sourced medical knowledge using the proposed multi-level graph contrastive learning framework. Moreover, The multi-channel sequence learning network facilitates capturing temporal relations between medical codes, enabling comprehensive patient representations for downstream tasks such as medication prediction.

Transcriptional regulation of KCS gene by bZIP29 and MYB70 transcription factors during ABA-stimulated wound suberization of kiwifruit (Actinidia deliciosa).

BACKGROUND: Our previous study has demonstrated that the transcription of AchnKCS involved in suberin biosynthesis was up-regulated by exogenous abscisic acid (ABA) during the wound suberization of kiwifruit, but the regulatory mechanism has not been fully elucidated. RESULTS: Through subcellular localization analysis in this work, AchnbZIP29 and AchnMYB70 transcription factors were observed to be localized in the nucleus. Yeast one-hybrid and dual-luciferase assay proved the transcriptional activation of AchnMYB70 and transcriptional suppression of AchnbZIP29 on AchnKCS promoter. Furthermore, the transcription level of AchnMYB70 was enhanced by ABA during wound suberization of kiwifruit, but AchnbZIP29 transcription was reduced by ABA. CONCLUSIONS: Therefore, it was believed that ABA enhanced the transcriptional activation of AchnMYB70 on AchnKCS by increasing AchnMYB70 expression. On the contrary, ABA relieved the inhibitory effect of AchnbZIP29 on transcription of AchnKCS by inhibiting AchnbZIP29 expression. These results gave further insight into the molecular regulatory network of ABA in wound suberization of kiwifruit.

Cross-Inhibitor: a time-sensitive molecular circuit based on DNA strand displacement.

Designing biochemical systems that can be effectively used in diverse fields, including diagnostics, molecular computing and nanomachines, has long been recognized as an important goal of molecular programming and DNA nanotechnology. A key issue in the development of such practical devices on the nanoscale lies in the development of biochemical components with information-processing capacity. In this article, we propose a molecular device that utilizes DNA strand displacement networks and allows interactive inhibition between two input signals; thus, it is termed a cross-inhibitor. More specifically, the device supplies each input signal with a processor such that the processing of one input signal will interdict the signal of the other. Biochemical experiments are conducted to analyze the interdiction performance with regard to effectiveness, stability and controllability. To illustrate its feasibility, a biochemical framework grounded in this mechanism is presented to determine the winner of a tic-tac-toe game. Our results highlight the potential for DNA strand displacement cascades to act as signal controllers and event triggers to endow molecular systems with the capability of controlling and detecting events and signals.

Effects of Lactobacillus plantarum and Saccharomyces cerevisiae co-fermentation on the structure and flavor of wheat noodles.

BACKGROUND: Although traditional fermented noodles possess high eating quality, it is difficult to realize large-scale industrialization as a result of the complexity of spontaneous fermentation. In present study, commercial Lactobacillus plantarum and Saccharomyces cerevisiae were applied in the preparation of fermented noodles. RESULTS: The changes in the structural characteristics and aroma components of noodles after fermentation were investigated via scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), low-field magenetic resonance imaging, electronic nose, and simultaneous distillation and extraction/gas chromatography-mass spectrometry (GC-MS) analysis. SEM images revealed that co-fermentation of the L. plantarum and S. cerevisiae for 10-40 min enhanced the continuity of the gluten network and promoted the formation of pores. FTIR spectra analysis showed that the co-fermentation increased significantly (P < 0.05) the proportion of α-helices of noodles gluten protein, enhancing the orderliness of the molecular structure of protein. After fermentation for 10-40 min, the signal density of hydrogen protons increased from the surface to the core, indicating that the water in the noodles migrated inward during a short fermentation process. The results of multivariate statistical analysis demonstrated that the main aroma differences between unfermented and fermented noodles were mainly in hydrocarbons, aromatic compounds and inorganic sulfides. GC-MS analysis indicated that the main volatile compounds detected were 2, 4-di-tert-butylphenol, bis (2-ethylhexyl) adipate, butyl acetate, dibutyl phthalate, dioctyl terephthalate, bis (2-ethylhexyl) phthalate, pentanol and 2-pentylfuran, etc. CONCLUSION: Co-fermentation with L. plantarum and S. cerevisiae improved the structure of gluten network and imparted more desirable volatile components to wheat noodles. © 2022 Society of Chemical Industry.

Allosteric DNAzyme-based DNA logic circuit: operations and dynamic analysis.

Recently, due to the dual roles of DNA and enzyme, DNAzyme has been widely used in the field of DNA circuit, which has a wide range of applications in bio-engineered system, information processing and biocomputing. In fact, the activity of DNAzymes was regulated by subunits assembly, pH control and metal ions triggers. However, those regulations required to change the sequences of whole DNAzyme, as separating parts and inserting extra DNA sequence. Inspired by the allosteric regulation of proteins in nature, a new allosteric strategy is proposed to regulate the activity of DNAzyme without DNA sequences changes. In this strategy, DNA strand displacement was used to regulate the DNAzyme structure, through which the activity of DNAzyme was well controlled. The strategy was applied to E6-type DNAzymes, and the operations of DNA logic circuit (YES, OR, AND, cascading and feedback) were established and simulated with the dynamic analyses. The allosteric regulation has potential to construct more complicated molecular systems, which can be applied to bio-sensing and detection.

Enriching limited information on rare diseases from heterogeneous networks for drug repositioning.

BACKGROUND: The historical data of rare disease is very scarce in reality, so how to perform drug repositioning for the rare disease is a great challenge. Most existing methods of drug repositioning for the rare disease usually neglect father-son information, so it is extremely difficult to predict drugs for the rare disease. METHOD: In this paper, we focus on father-son information mining for the rare disease. We propose GRU-Cooperation-Attention-Network (GCAN) to predict drugs for the rare disease. We construct two heterogeneous networks for information enhancement, one network contains the father-nodes of the rare disease and the other network contains the son-nodes information. To bridge two heterogeneous networks, we set a mapping to connect them. What's more, we use the biased random walk mechanism to collect the information smoothly from two heterogeneous networks, and employ a cooperation attention mechanism to enhance repositioning ability of the network. RESULT: Comparing with traditional methods, GCAN makes full use of father-son information. The experimental results on real drug data from hospitals show that GCAN outperforms state-of-the-art machine learning methods for drug repositioning. CONCLUSION: The performance of GCAN for drug repositioning is mainly limited by the insufficient scale and poor quality of the data. In future research work, we will focus on how to utilize more data such as drug molecule information and protein molecule information for the drug repositioning of the rare disease.

ABF2 and MYB transcription factors regulate feruloyl transferase FHT involved in ABA-mediated wound suberization of kiwifruit.

Suberin is a cell-wall biopolymer with aliphatic and aromatic domains that is synthesized in the wound tissues of plants in order to restrict water loss and pathogen infection. ω-hydroxyacid/fatty alcohol hydroxycinnamoyl transferase (FHT) is required for cross-linking of the aliphatic and aromatic domains. ABA is known to play a positive role in suberin biosynthesis but it is not known how it interacts with FHT. In this study, the kiwifruit (Actinidia chinensis) AchnFHT gene was isolated and was found to be localized in the cytosol. Transient overexpression of AchnFHT in leaves of Nicotiana benthamiana induced massive production of ferulate, ω-hydroxyacids, and primary alcohols, consistent with the in vitro ability of AchnFHT to catalyse acyl-transfer from feruloyl-CoA to ω-hydroxypalmitic acid and 1-tetradecanol. A regulatory function of four TFs (AchnABF2, AchnMYB4, AchnMYB41, and AchnMYB107) on AchnFHT was identified. These TFs localized in the nucleus and directly interacted with the AchnFHT promoter in yeast one-hybrid assays. Dual-luciferase analysis indicated that AchnABF2, AchnMYB41, and AchnMYB107 activated the AchnFHT promoter while AchnMYB4 repressed it. These findings were supported by the results of transient overexpression in N. benthamiana, in which AchnABF2, AchnMYB41, and AchnMYB107 induced expression of suberin biosynthesis genes (including FHT) and accumulation of suberin monomers, whilst AchnMYB4 had the opposite effect. Exogenous ABA induced the expression of AchnABF2, AchnMYB41, AchnMYB107, and AchnFHT and induced suberin monomer formation, but it inhibited AchnMYB4 expression. In addition, fluridone (an inhibitor of ABA biosynthesis) was found to counter the inductive effects of ABA. Activation of suberin monomer biosynthesis by AchnFHT was therefore controlled in a coordinated way by both repression of AchnMYB4 and promotion of AchnABF2, AchnMYB41, and AchnMYB107.

A DNAzyme-mediated logic gate system based on Ag(I)-cysteine.

Ag+ plays an important role in DNA mismatch technology due to its affinity for cytosine in DNA. This article introduces a strategy to control the enzyme digesting reaction by utilizing the characteristics of C-Ag+-C mismatches, effectively regulating and controlling the activity of the E6 DNAzyme via changing the structure of its conservative domain. We designed a series of basic logic gates, a "Yes" Gate, an "Or" Gate and an "Inhibit" Gate. Cysteine (Cys) can combine with Ag+, reducing the concentration of Ag+ in solution, thus restraining the C-Ag+-C mismatch effect. Based on this principle, we regard Cys as a threshold, and designed a type of "Inhibit" Gate based on input quantity by changing the concentration of Ag+, thus generating different statues of logic output. On this basis, the E6 DNAzyme and Ag10c DNAzyme can be integrated into new systems with the function of logic operation circuit based on the control of Ag+ concentration in solution. This system could represent three different states of logical expression by controlling the quantity of Ag+ and Cys.

RAHM: Relation augmented hierarchical multi-task learning framework for reasonable medication stocking.

As an important task in digital preventive healthcare management, especially in the secondary prevention stage, active medication stocking refers to the process of preparing necessary medications in advance according to the predicted disease progression of patients. However, predicting preventive or even life-saving medicine for each patient is a non-trivial task. Existing models usually overlook the implicit hierarchical relation between patient's predicted diseases and medications, and mainly focus on single tasks (medication recommendation or disease prediction). To tackle this limitation, we propose a relation augmented hierarchical multi-task learning framework, named RAHM. which is capable of learning multi-level relation-aware patient representation for reasonable medication stocking. Specifically, the framework first leverages the underlying structural relations of Electronic Health Record (EHR) data to learn the low-level patient visit representation. Then, it uses a regular LSTM to encode the historical temporal disease information for disease-level patient representation learning. Further, a relation-aware LSTM (R-LSTM) is proposed to handle the relations between diseases and medication in longitudinal patient records, which can better integrate the historical information into the medication-level patient representation. In the learning process, two pseudo residual structures are introduced to mitigate the error propagation and preserve the valuable relation information of EHRs. To validate our method, extensive experiments have been conducted based on the real-world clinical dataset. The results demonstrate a consistent superiority of our framework over several baselines in suggesting reasonable stock medication.

PASCAL: a pseudo cascade learning framework for breast cancer treatment entity normalization in Chinese clinical text.

BACKGROUNDS: Knowledge discovery from breast cancer treatment records has promoted downstream clinical studies such as careflow mining and therapy analysis. However, the clinical treatment text from electronic health data might be recorded by different doctors under their hospital guidelines, making the final data rich in author- and domain-specific idiosyncrasies. Therefore, breast cancer treatment entity normalization becomes an essential task for the above downstream clinical studies. The latest studies have demonstrated the superiority of deep learning methods in named entity normalization tasks. Fundamentally, most existing approaches adopt pipeline implementations that treat it as an independent process after named entity recognition, which can propagate errors to later tasks. In addition, despite its importance in clinical and translational research, few studies directly deal with the normalization task in Chinese clinical text due to the complexity of composition forms. METHODS: To address these issues, we propose PASCAL, an end-to-end and accurate framework for breast cancer treatment entity normalization (TEN). PASCAL leverages a gated convolutional neural network to obtain a representation vector that can capture contextual features and long-term dependencies. Additionally, it treats treatment entity recognition (TER) as an auxiliary task that can provide meaningful information to the primary TEN task and as a particular regularization to further optimize the shared parameters. Finally, by concatenating the context-aware vector and probabilistic distribution vector from TEN, we utilize the conditional random field layer (CRF) to model the normalization sequence and predict the TEN sequential results. RESULTS: To evaluate the effectiveness of the proposed framework, we employ the three latest sequential models as baselines and build the model in single- and multitask on a real-world database. Experimental results show that our method achieves better accuracy and efficiency than state-of-the-art approaches. CONCLUSIONS: The effectiveness and efficiency of the presented pseudo cascade learning framework were validated for breast cancer treatment normalization in clinical text. We believe the predominant performance lies in its ability to extract valuable information from unstructured text data, which will significantly contribute to downstream tasks, such as treatment recommendations, breast cancer staging and careflow mining.

An Intelligent Optimization Algorithm for Constructing a DNA Storage Code: NOL-HHO.

The high density, large capacity, and long-term stability of DNA molecules make them an emerging storage medium that is especially suitable for the long-term storage of large datasets. The DNA sequences used in storage need to consider relevant constraints to avoid nonspecific hybridization reactions, such as the No-runlength constraint, GC-content, and the Hamming distance. In this work, a new nonlinear control parameter strategy and a random opposition-based learning strategy were used to improve the Harris hawks optimization algorithm (for the improved algorithm NOL-HHO) in order to prevent it from falling into local optima. Experimental testing was performed on 23 widely used benchmark functions, and the proposed algorithm was used to obtain better coding lower bounds for DNA storage. The results show that our algorithm can better maintain a smooth transition between exploration and exploitation and has stronger global exploration capabilities as compared with other algorithms. At the same time, the improvement of the lower bound directly affects the storage capacity and code rate, which promotes the further development of DNA storage technology.

High oxygen facilitates wound induction of suberin polyphenolics in kiwifruit.

BACKGROUND: Rapid wound healing would be critical for successful long-term storage of fruits and vegetables. However, there was no direct evidence for the requirement and efficiency of oxygen in the fruit wound-healing process. This study was conducted to investigate the role of oxygen in wound-induced suberization by analyzing melanin, suberin polyphenolics (SPPs) and related enzymes in half-cut kiwifruits exposed to 100%, 50%, 21% and 0% oxygen. RESULTS: By 3 days after wounding, the wound surface of kiwifruit in high (50 and 100%) oxygen appeared as a continuous layer of melanin and SPPs underneath, which effectively prevent excessive water vapor loss from the fruit halves. In contrast, melanin and SPPs deposition in the wound surface in 0% oxygen was significantly reduced, with high water vapor loss. Rapid decrease of soluble phenolic acids (caffeic, p-coumaric, ferulic acids) was coupled with the increase of bound ferulic acid (coniferyl diacetate) especially in high oxygen by 9 days after wounding. Meanwhile, high oxygen enhanced peroxidase, catalase, phenylalanine ammonia-lyase, and polyphenol oxidase activities. CONCLUSION: Oxygen is required for wound-induced melanin and SPPs formation, and high oxygen is effective in promoting wound suberization in postharvest kiwifruit. © 2017 Society of Chemical Industry.

Dual modality prompt learning for visual question-grounded answering in robotic surgery.

With recent advancements in robotic surgery, notable strides have been made in visual question answering (VQA). Existing VQA systems typically generate textual answers to questions but fail to indicate the location of the relevant content within the image. This limitation restricts the interpretative capacity of the VQA models and their ability to explore specific image regions. To address this issue, this study proposes a grounded VQA model for robotic surgery, capable of localizing a specific region during answer prediction. Drawing inspiration from prompt learning in language models, a dual-modality prompt model was developed to enhance precise multimodal information interactions. Specifically, two complementary prompters were introduced to effectively integrate visual and textual prompts into the encoding process of the model. A visual complementary prompter merges visual prompt knowledge with visual information features to guide accurate localization. The textual complementary prompter aligns visual information with textual prompt knowledge and textual information, guiding textual information towards a more accurate inference of the answer. Additionally, a multiple iterative fusion strategy was adopted for comprehensive answer reasoning, to ensure high-quality generation of textual and grounded answers. The experimental results validate the effectiveness of the model, demonstrating its superiority over existing methods on the EndoVis-18 and EndoVis-17 datasets.

Active substance and mechanisms of Actinidia chinensis Planch for the treatment of breast cancer was explored based on network pharmacology and in silico study.

In this paper, our objective was to investigate the potential mechanisms of Actinidia chinensis Planch (ACP) for breast cancer treatment with the application of network pharmacology, molecular docking, and molecular dynamics. "Mihoutaogen" was used as a key word to query the Traditional Chinese Medicine Systems Pharmacology database for putative ingredients of ACP and its related targets. DrugBank, GeneCards, Online Mendelian Inheritance in Man, and therapeutic target databases were used to search for genes associated with "breast cancer." Using Cytoscape 3.9.0 we then constructed the protein-protein interaction and drug-ingredient-target-disease networks. An enrichment analysis of Kyoto encyclopedia of genes and genomes pathway and gene ontology were performed to exploration of the signaling pathways associated with ACP for breast cancer treatment. Discovery Studio software was applied to molecular docking. Finally, the ligand-receptor complex was subjected to a 50-ns molecular dynamics simulation using the Desmond_2020.4 tools. Six main active ingredients and 176 targets of ACP and 2243 targets of breast cancer were screened. There were 118 intersections of targets for both active ingredients and diseases. Tumor protein P53 (TP53), AKT serine/threonine kinase 1 (AKT1), estrogen receptor 1 (ESR1), Erb-B2 receptor tyrosine kinase 2 (ERBB2), epidermal growth factor receptor (EGFR), Jun Proto-Oncogene (JUN), and Heat Shock Protein 90 Alpha Family Class A Member 1 (HSP90AA1) selected as the most important genes were used for verification by molecular docking and molecular dynamics simulation. The primary active compounds of ACP against breast cancer were predicted preliminarily, and its mechanism was studied, thereby providing a theoretical basis for future clinical studies.

Eye Gaze Guided Cross-Modal Alignment Network for Radiology Report Generation.

The potential benefits of automatic radiology report generation, such as reducing misdiagnosis rates and enhancing clinical diagnosis efficiency, are significant. However, existing data-driven methods lack essential medical prior knowledge, which hampers their performance. Moreover, establishing global correspondences between radiology images and related reports, while achieving local alignments between images correlated with prior knowledge and text, remains a challenging task. To address these shortcomings, we introduce a novel Eye Gaze Guided Cross-modal Alignment Network (EGGCA-Net) for generating accurate medical reports. Our approach incorporates prior knowledge from radiologists' Eye Gaze Region (EGR) to refine the fidelity and comprehensibility of report generation. Specifically, we design a Dual Fine-Grained Branch (DFGB) and a Multi-Task Branch (MTB) to collaboratively ensure the alignment of visual and textual semantics across multiple levels. To establish fine-grained alignment between EGR-related images and sentences, we introduce the Sentence Fine-grained Prototype Module (SFPM) within DFGB to capture cross-modal information at different levels. Additionally, to learn the alignment of EGR-related image topics, we introduce the Multi-task Feature Fusion Module (MFFM) within MTB to refine the encoder output information. Finally, a specifically designed label matching mechanism is designed to generate reports that are consistent with the anticipated disease states. The experimental outcomes indicate that the introduced methodology surpasses previous advanced techniques, yielding enhanced performance on two extensively used benchmark datasets: Open-i and MIMIC-CXR.

Bubble DNA tweezer: A triple-conformation sensor responsive to concentration-ratios.

DNA tweezers, with their elegant simplicity and flexibility, have been pivotal in biosensing and DNA computing. However, conventional tweezers are confined to a binary transformation pre/post target signal recognition, limiting them to presence/absence judgments. This study introduces bubble DNA tweezers (BDT), capable of three distinct conformations based on variable target signal ratios. In contrast to traditional compact tweezers, BDT features a looser structure centered around a non-complementary bubble domain located between the tweezer arms' connecting axis and target signal recognition jaws. This bubble triggers toehold-free DNA strand displacement, leading to three conformational changes at different target signal concentrations. BDT detects presence/absence and true concentration with remarkable specificity and sensitivity. This adaptability is not confined to ideal scenarios, proving valuable in complex, noisy environments. Our method facilitates target DNA/miRNA signal quantification within a specific length range, promising applications in clinical research and environmental detection, while inspiring future biological assay innovations.