Disambiguity and Alignment: An Effective Multi-Modal Alignment Method for Cross-Modal Recipe Retrieval.

As a prominent topic in food computing, cross-modal recipe retrieval has garnered substantial attention. However, the semantic alignment across food images and recipes cannot be further enhanced due to the lack of intra-modal alignment in existing solutions. Additionally, a critical issue named food image ambiguity is overlooked, which disrupts the convergence of models. To these ends, we propose a novel Multi-Modal Alignment Method for Cross-Modal Recipe Retrieval (MMACMR). To consider inter-modal and intra-modal alignment together, this method measures the ambiguous food image similarity under the guidance of their corresponding recipes. Additionally, we enhance recipe semantic representation learning by involving a cross-attention module between ingredients and instructions, which is effective in supporting food image similarity measurement. We conduct experiments on the challenging public dataset Recipe1M; as a result, our method outperforms several state-of-the-art methods in commonly used evaluation criteria.

DRANetSplicer: A Splice Site Prediction Model Based on Deep Residual Attention Networks.

The precise identification of splice sites is essential for unraveling the structure and function of genes, constituting a pivotal step in the gene annotation process. In this study, we developed a novel deep learning model, DRANetSplicer, that integrates residual learning and attention mechanisms for enhanced accuracy in capturing the intricate features of splice sites. We constructed multiple datasets using the most recent versions of genomic data from three different organisms, Oryza sativa japonica, Arabidopsis thaliana and Homo sapiens. This approach allows us to train models with a richer set of high-quality data. DRANetSplicer outperformed benchmark methods on donor and acceptor splice site datasets, achieving an average accuracy of (96.57%, 95.82%) across the three organisms. Comparative analyses with benchmark methods, including SpliceFinder, Splice2Deep, Deep Splicer, EnsembleSplice, and DNABERT, revealed DRANetSplicer's superior predictive performance, resulting in at least a (4.2%, 11.6%) relative reduction in average error rate. We utilized the DRANetSplicer model trained on O. sativa japonica data to predict splice sites in A. thaliana, achieving accuracies for donor and acceptor sites of (94.89%, 94.25%). These results indicate that DRANetSplicer possesses excellent cross-organism predictive capabilities, with its performance in cross-organism predictions even surpassing that of benchmark methods in non-cross-organism predictions. Cross-organism validation showcased DRANetSplicer's excellence in predicting splice sites across similar organisms, supporting its applicability in gene annotation for understudied organisms. We employed multiple methods to visualize the decision-making process of the model. The visualization results indicate that DRANetSplicer can learn and interpret well-known biological features, further validating its overall performance. Our study systematically examined and confirmed the predictive ability of DRANetSplicer from various levels and perspectives, indicating that its practical application in gene annotation is justified.

Flavonoids from mulberry leaves inhibit fat production and improve fatty acid distribution in adipose tissue in finishing pigs.

This study evaluated the effects of flavonoids from mulberry leaves (FML) on plasma biochemical indices, serum activities of lipid metabolism-related enzymes, fat morphology, fatty acid composition, and lipid metabolism in different adipose tissues of finishing pigs. We used 120 Chinese hybrid barrows of Berkshire and Bama mini-pigs with an average initial body weight of 45.11 ± 4.23 kg. The pigs were randomly assigned to five treatment groups and fed a control diet based on corn, soybean meal, and wheat bran or a control diet supplemented with 0.02%, 0.04%, 0.08%, or 0.16% FML. Each experimental group had six replicates (pens), with four pigs per pen. After a 7-d adaptation period, the feeding trial was conducted for 58 d. Blood and adipose tissue samples were collected from 30 pigs (one pig per pen) at the end of the test. The results showed that FML supplementation significantly decreased the feed intake to body gain ratio, the plasma concentrations of total cholesterol and free fatty acids, and the serum activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase (linear or quadratic effects, P < 0.05), and decreased the plasma triglyceride concentration (quadratic, P = 0.07). Increasing FML supplementation increased the average daily gain and serum activities of lipoprotein lipase (linear and quadratic effects, P < 0.05) and adipose triglyceride lipase (linear, P < 0.05). Dietary FML supplementation decreased the adipocyte area in the dorsal subcutaneous adipose (DSA) tissue of finishing pigs (linear, P = 0.05) and increased the adipocyte area in the visceral adipose tissue (quadratic, P < 0.01). Increasing FML supplementation decreased the C20:1 content in DSA, abdominal subcutaneous adipose, and visceral adipose tissues of finishing pigs (P < 0.05) and increased the C18:3n3 and n-3 PUFA contents (P < 0.05). The lipid metabolism genes were regulated by the PPARγ-LXRα-ABCA1 signaling pathway, and their expressions differed in different adipose tissues. These findings suggest that FML improved growth performance, regulated lipid metabolism, inhibited fat production, and improved fatty acid distribution in the adipose tissue of finishing pigs, thereby improving pig fat's nutritional quality and health value.

Spatial-Spectral Constrained Adaptive Graph for Hyperspectral Image Clustering.

Hyperspectral image (HSI) clustering is a challenging task, whose purpose is to assign each pixel to a corresponding cluster. The high-dimensionality and noise corruption are two main problems that limit the performance of HSI clustering. To address those problems, this paper proposes a projected clustering with a spatial-spectral constrained adaptive graph (PCSSCAG) method for HSI clustering. PCSSCAG first constructs an adaptive adjacency graph to capture the accurate local geometric structure of HSI data adaptively. Then, a spatial-spectral constraint is employed to simultaneously explore the spatial and spectral information for reducing the negative influence on graph construction caused by noise. Finally, projection learning is integrated into the spatial-spectral constrained adaptive graph construction for reducing the redundancy and alleviating the computational cost. In addition, an alternating iteration algorithm is designed to solve the proposed model, and its computational complexity is theoretically analyzed. Experiments on two different scales of HSI datasets are conducted to evaluate the performance of PCSSCAG. The associated experimental results demonstrate the superiority of the proposed method for HSI clustering.

Efficacy and safety of probiotics in Parkinson's constipation: A systematic review and meta-analysis.

Background: Parkinson's disease (PD) is the most common neurodegenerative disease closely related to the immune system, among whose prodromes constipation is a representative symptom. Recent Randomized Controlled Trials (RCTs) have proved that probiotics can be used to effectively treat PD constipation, but the results are inconsistent. We performed a meta-analysis to assess the efficacy and safety of probiotic therapy on Parkinson's constipation. Methods: Questions about the research focus were constructed based on the Participants, Intervention, Comparison and Outcomes (PICO) Criteria. We searched electronic databases such as PubMed, Web of Science, EMBASE, Scopus, EBSCO, Cochrane and Google Scholar until March 2022 for eligible literatures. Our primary endpoints were stool frequency, stool consistency, the number of laxatives uses, UPDRS-III scores and adverse events. Results: 12 eligible studies (n = 818 patients) met the inclusion and endpoint criteria. Meta-analysis results showed that constipation symptoms were improved after probiotic treatment, including an increased stool frequency (WMD = 0.94, 95% CI:0.53 to 1.34; OR = 3.22, 95% CI:1.97-5.29), an improved stool consistency (WMD = 1.46, 95% CI:0.54-2.37), a reduced use of laxatives (WMD = -0.72, 95%CI: -1.04 to-0.41), and also a reduced Parkinson's UPDRS-III score (WMD = -6.58, 95%CI: -12.02 to -1.14); there was no significant difference in total adverse events (OR = 0.82, 95%CI:0.39-1.72). Conclusion: Our analysis suggests that probiotics can be used to improve the constipation and motor symptoms for patients with Parkinson's constipation, possibly by reducing the inflammatory response and improving gut-brain axis neuron function, whose safety also proved to be good.

Effectiveness and safety of warm needle acupuncture on lumbar muscles strain: A protocol for systematic review and meta analysis.

BACKGROUND: Lumbar muscle strain (LMS) is the most common orthopedic syndrome, with high incidence globally and lingering disease, which seriously affects patients' work efficiency and quality of life. Warm needle acupuncture (WNA) is a treatment method combining acupuncture technology with warm and medicinal effect of moxibustion. It has outstanding curative effect and wide range of treatment, especially in the treatment of pain diseases. We aim to collect clinical evidence and demonstrate the efficacy and safety of WNA on LMS. METHODS/DESIGN: We will search the following database sources for the randomized controlled trials: PubMed, Cochrane Library, Excerpta Medica Database (EMBASE), Web of Science, WHO International Clinical Trials Registry Platform (TCTRP), Chinese Biomedical Literature Database (CBM), Chinese National Knowledge Infrastructure Database (CNKI), Chinese Scientific Journals Database (VIP), and the Wanfang Database.All randomized controlled trials of WNA for lumbar muscle strain (LMS) in the above database will be considered for inclusion, and high-quality articles will be screened for data extraction and analysis, to summarize the therapeutic effect of WNA on LMS patients. RESULT: This study will provide a rational synthesis of current evidences for warm needle acupuncture on lumbar muscle strain. CONCLUSION: The conclusion of this study will provide evidence to judge the effectiveness and safety of WNA on LMS. TRIAL REGISTRATION: INPLASY2020120100 (DOI number: 10.37766/inplasy2020.12.0100).

Therapy of diarrhea in COVID-19 with external treatment of traditional Chinese medicine: A protocol for systematic review and meta-analysis.

BACKGROUND: Evaluating the effectiveness and safety of external treatment of traditional Chinese medicine therapy for COVID-19 with diarrhea is the primary purpose of this systematic evaluation program. METHODS: We will search the randomized controlled trials from inception to November 2020. The following database is our focus area: Cochrane Central Register of Controlled Trials, Embase, PubMed, Web of Science, China National Knowledge Infrastructure (CNKI), Chinese Biomedical Database (CBM), VIP database for Chinese technical Periodicals, and Wanfang Database. We will choose articles published both in Chinese and English. Two reviewers will conduct the study selection, data extraction, and assessment independently. The assessment of risk of bias and data synthesis will be carried out using Review Manager Software V.5.3. RESULTS: The results will provide high-quality synthesis of current evidence for researchers in this subject area. CONCLUSION: This studys decision will provide evidence of whether external treatment of traditional Chinese medicine is an effective and safe intervention for coronavirus disease 2019 with diarrhea. REGISTRATION NUMBER: INPLASY2020110095 (DOI number: 10.37766/inplasy2020.11.0095).

Digital image modification detection using color information and its histograms.

The rapid development of many open source and commercial image editing software makes the authenticity of the digital images questionable. Copy-move forgery is one of the most widely used tampering techniques to create desirable objects or conceal undesirable objects in a scene. Existing techniques reported in the literature to detect such tampering aim to improve the robustness of these methods against the use of JPEG compression, blurring, noise, or other types of post processing operations. These post processing operations are frequently used with the intention to conceal tampering and reduce tampering clues. A robust method based on the color moments and other five image descriptors is proposed in this paper. The method divides the image into fixed size overlapping blocks. Clustering operation divides entire search space into smaller pieces with similar color distribution. Blocks from the tampered regions will reside within the same cluster since both copied and moved regions have similar color distributions. Five image descriptors are used to extract block features, which makes the method more robust to post processing operations. An ensemble of deep compositional pattern-producing neural networks are trained with these extracted features. Similarity among feature vectors in clusters indicates possible forged regions. Experimental results show that the proposed method can detect copy-move forgery even if an image was distorted by gamma correction, addictive white Gaussian noise, JPEG compression, or blurring.