Panoramic imaging errors in machine learning model development: a systematic review.

OBJECTIVES: To investigate the management of imaging errors from panoramic radiography (PAN) datasets used in the development of machine learning (ML) models. METHODS: This systematic literature followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and used three databases. Keywords were selected from relevant literature. ELIGIBILITY CRITERIA: PAN studies that used ML models and mentioned image quality concerns. RESULTS: Out of 400 articles, 41 papers satisfied the inclusion criteria. All the studies used ML models, with 35 papers using deep learning (DL) models. PAN quality assessment was approached in 3 ways: acknowledgement and acceptance of imaging errors in the ML model, removal of low-quality radiographs from the dataset before building the model, and application of image enhancement methods prior to model development. The criteria for determining PAN image quality varied widely across studies and were prone to bias. CONCLUSIONS: This study revealed significant inconsistencies in the management of PAN imaging errors in ML research. However, most studies agree that such errors are detrimental when building ML models. More research is needed to understand the impact of low-quality inputs on model performance. Prospective studies may streamline image quality assessment by leveraging DL models, which excel at pattern recognition tasks.

Exploring the Roles of Single Atom in Hydrogen Peroxide Photosynthesis.

This comprehensive review provides a deep exploration of the unique roles of single atom catalysts (SACs) in photocatalytic hydrogen peroxide (H2O2) production. SACs offer multiple benefits over traditional catalysts such as improved efficiency, selectivity, and flexibility due to their distinct electronic structure and unique properties. The review discusses the critical elements in the design of SACs, including the choice of metal atom, host material, and coordination environment, and how these elements impact the catalytic activity. The role of single atoms in photocatalytic H2O2 production is also analysed, focusing on enhancing light absorption and charge generation, improving the migration and separation of charge carriers, and lowering the energy barrier of adsorption and activation of reactants. Despite these advantages, several challenges, including H2O2 decomposition, stability of SACs, unclear mechanism, and low selectivity, need to be overcome. Looking towards the future, the review suggests promising research directions such as direct utilization of H2O2, high-throughput synthesis and screening, the creation of dual active sites, and employing density functional theory for investigating the mechanisms of SACs in H2O2 photosynthesis. This review provides valuable insights into the potential of single atom catalysts for advancing the field of photocatalytic H2O2 production.

Regulation of HHLA2 expression in kidney cancer and myeloid cells.

BACKGROUND: The immune checkpoint HERV-H LTR-associating 2 (HHLA2) is expressed in kidney cancer and various other tumor types. Therapeutics targeting HHLA2 or its inhibitory receptor KIR3DL3 are being developed for solid tumors, including renal cell carcinoma (RCC). However, the regulation of HHLA2 expression remains poorly understood. A better understanding of HHLA2 regulation in tumor cells and the tumor microenvironment is crucial for the successful translation of these therapeutic agents into clinical applications. METHODS: Flow cytometry and quantitative real-time PCR were used to analyze HHLA2 expression in primary kidney tumors ex vivo and during in vitro culture. HHLA2 expression in A498 and 786-O ccRCC cell lines was examined in vitro and in subcutaneous tumor xenografts in NSG mice. Monocytes and dendritic cells were analyzed for HHLA2 expression. We tested a range of cytokines and culture conditions, including hypoxia, to induce HHLA2 expression. RESULTS: Analysis of HHLA2 expression revealed that HHLA2 is expressed on tumor cells in primary kidney tumors ex vivo; however, its expression gradually diminishes during a 4-week in vitro culture period. A498 and 786-O ccRCC tumor cell lines do not express HHLA2 in vitro, but HHLA2 expression was observed when grown as subcutaneous xenografts in NSG immunodeficient mice. Induction experiments using various cytokines and culture conditions failed to induce HHLA2 expression in A498 and 786-O tumor cell lines in vitro. Analysis of HHLA2 expression in monocytes and dendritic cells demonstrated that only IL-10 and BMP4, along with IL-1ß and IL-6 to a lesser extent, modestly enhanced HHLA2 protein and mRNA expression. CONCLUSIONS: HHLA2 expression is induced on kidney cancer cells in vivo by a tumor microenvironmental signal that is not present in vitro. HHLA2 expression is differentially regulated in kidney cancer epithelial cells and monocytes. Cytokines, particularly IL10, that induce HHLA2 expression in monocytes fail to upregulate HHLA2 expression in tumor cell lines in vitro. These findings underscore the importance of the interplay between tumor cell and tumor microenvironmental signals in the regulation of HHLA2. Further investigation is warranted to elucidate the mechanisms involved in HHLA2 regulation and its implications for therapeutic development.

scGGAN: single-cell RNA-seq imputation by graph-based generative adversarial network.

Single-cell RNA sequencing (scRNA-seq) data are typically with a large number of missing values, which often results in the loss of critical gene signaling information and seriously limit the downstream analysis. Deep learning-based imputation methods often can better handle scRNA-seq data than shallow ones, but most of them do not consider the inherent relations between genes, and the expression of a gene is often regulated by other genes. Therefore, it is essential to impute scRNA-seq data by considering the regional gene-to-gene relations. We propose a novel model (named scGGAN) to impute scRNA-seq data that learns the gene-to-gene relations by Graph Convolutional Networks (GCN) and global scRNA-seq data distribution by Generative Adversarial Networks (GAN). scGGAN first leverages single-cell and bulk genomics data to explore inherent relations between genes and builds a more compact gene relation network to jointly capture the homogeneous and heterogeneous information. Then, it constructs a GCN-based GAN model to integrate the scRNA-seq, gene sequencing data and gene relation network for generating scRNA-seq data, and trains the model through adversarial learning. Finally, it utilizes data generated by the trained GCN-based GAN model to impute scRNA-seq data. Experiments on simulated and real scRNA-seq datasets show that scGGAN can effectively identify dropout events, recover the biologically meaningful expressions, determine subcellular states and types, improve the differential expression analysis and temporal dynamics analysis. Ablation experiments confirm that both the gene relation network and gene sequence data help the imputation of scRNA-seq data.

Differential Gene Expression Prediction by Ensemble Deep Networks on Histone Modification Data.

Predicting differential gene expression (DGE) from Histone modifications (HM) signal is crucial to understand how HM controls cell functional heterogeneity through influencing differential gene regulation. Most existing prediction methods use fixed-length bins to represent HM signals and transmit these bins into a single machine learning model to predict differential expression genes of single cell type or cell type pair. However, the inappropriate bin length may cause the splitting of the important HM segment and lead to information loss. Furthermore, the bias of single learning model may limit the prediction accuracy. Considering these problems, in this paper, we proposes an Ensemble deep neural networks framework for predicting Differential Gene Expression (EnDGE). EnDGE employs different feature extractors on input HM signal data with different bin lengths and fuses the feature vectors for DGE prediction. Ensemble multiple learning models with different HM signal cutting strategies helps to keep the integrity and consistency of genetic information in each signal segment, and offset the bias of individual models. Besides the popular feature extractors, we also propose a new Residual Network based model with higher prediction accuracy to increase the diversity of feature extractors. Experiments on the real datasets from the Roadmap Epigenome Project (REMC) show that for all cell type pairs, EnDGE significantly outperforms the state-of-the-art baselines for differential gene expression prediction.

Transcriptome analysis of differentially expressed genes involved in the inflammageing status of gingiva in aged mice.

OBJECTIVE: To identify immune-inflammation-related genes related to susceptibility to periodontitis in the gingiva of aged mice with RNA sequencing. METHODS: Gingival samples from 18-month-old, 8-week-old healthy mice and 8-week-old mice with periodontitis were taken for RNA-seq. The differentially expressed genes (DEGs) were validated with qRT-PCR using mouse and human gingival samples. RESULTS: 977 (upregulated) and 1824 (downregulated) genes were identified in the old compared with the young mice. 14.2% were related to immune-inflammatory responses. This proportion of overlap (ageing and periodontitis)-DEGs was higher (48.4%). Enrichment analysis of overlap (ageing and periodontitis)-DEGs showed that the top five GO and KEGG terms were related to the immune-inflammatory responses, and disease analysis was more specific to periodontitis. The candidate genes of overlap (ageing and periodontitis)-DEGs selected by protein-protein interaction (PPI) network showed the higher match with clinical data sets. By qRT-PCR, nine candidate genes were identified as hub genes that are associated with susceptibility to periodontitis in the elderly, including CXCL3, CXCL5, CSF3, CSF3R, FPR1, IL1B, OSM, SERPINE1 and SELP. CONCLUSION: Our studies provide insights into the mechanisms by which ageing affects the immune-inflammatory status of gingival tissues, thereby increasing the risk of periodontitis. It may become targets for future prevention of periodontitis.

Ultrastable Zinc Anode Enabled by CO2-Induced Interface Layer.

Aqueous Zn-ion batteries (AZIBs), being safe, inexpensive, and pollution-free, are a promising candidate for future large-scale sustainable energy storage. However, in a conventional AZIBs setup, the Zn metal anode suffers oxidative corrosion, side reactions with electrolytes, disordered dendrite growth during operation, and consequently low efficiency and short lifespan. In this work, we discover that purging CO2 gas into the electrolyte could address these issues by eliminating dissolved O2, inhibiting side reactions by buffering the local pH change, and preventing dendrite growth by inducing the in situ formation of a ZnCO3 solid electrolyte interphase layer. Moreover, the CO2-purged electrolyte could enable a highly reversible plating/stripping behavior with a high Coulombic efficiency of 99.97% and an ultralong lifespan of 32,000 cycles (1600 h) even under an ultrahigh current density of 40 mA cm-2. Consequently, the CO2-purged symmetrical cells deliver long cycling stability at a high depth of discharge of 57%, while the CO2-purged Zn/V2O5 full cells exhibit outstanding capacity retention of 66% after 1000 cycles at a high current density of 5 A g-1. Our strategy, the simple introduction of CO2 gas into the electrolyte, could effectively mediate the zinc anode's critical issues and provide a scalable and cost-effective pathway for the commercialization of AZIBs.

Differentially expressed genes prediction by multiple self-attention on epigenetics data.

Predicting differentially expressed genes (DEGs) from epigenetics signal data is the key to understand how epigenetics controls cell functional heterogeneity by gene regulation. This knowledge can help developing 'epigenetics drugs' for complex diseases like cancers. Most of existing machine learning-based methods suffer defects in prediction accuracy, interpretability or training speed. To address these problems, in this paper, we propose a Multiple Self-Attention model for predicting DEGs on Epigenetic data (Epi-MSA). Epi-MSA first uses convolutional neural networks for neighborhood bins information embedding, and then employs multiple self-attention encoders on different input epigenetics factors data to learn which locations of genes are important for predicting DEGs. Next it trains a soft attention module to pick out which epigenetics factors are significant. The attention mechanism makes the model interpretable, and the pure matrix operation of self-attention enables the model to be parallel calculated and speeds up the training. Experiments on datasets from the Roadmap Epigenome Project and BluePrint Data Analysis Portal (BDAP) show that the performance of Epi-MSA is better than existing competitive methods, and Epi-MSA also has a smaller standard deviation, which shows that Epi-MSA is effective and stable. In addition, Epi-MSA has a good interpretability, this is confirmed by referring its attention weight matrix with existing biological knowledge.

Combining CNN with Pathological Information for the Detection of Transmissive Lesions of Jawbones from CBCT Images.

Cone-Beam Computed Tomography (CBCT) imaging modality is used to acquire 3D volumetric image of the human body. CBCT plays a vital role in diagnosing dental diseases, especially cyst or tumour-like lesions. Current computer-aided detection and diagnostic systems have demonstrated diagnostic value in a range of diseases, however, the capability of such a deep learning method on transmissive lesions has not been investigated. In this study, we propose an automatic method for the detection of transmissive lesions of jawbones using CBCT images. We integrated a pre-trained DenseNet with pathological information to reduce the intra-class variation within a patient's images in the 3D volume (stack) that may affect the performance of the model. Our proposed method separates each CBCT stacks into seven intervals based on their disease manifestation. To evaluate the performance of our method, we created a new dataset containing 353 patients' CBCT data. A patient-wise image division strategy was employed to split the training and test sets. The overall lesion detection accuracy of 80.49% was achieved, outperforming the baseline DenseNet result of 77.18%. The result demonstrates the feasibility of our method for detecting transmissive lesions in CBCT images.Clinical relevance - The proposed strategy aims at providing automatic detection of the transmissive lesions of jawbones with the use of CBCT images that can reduce the workload of clinical radiologists, improve their diagnostic efficiency, and meet the preliminary requirement for the diagnosis of this kind of disease when there is a lack of radiologists.

The mental health of transgender and gender non-conforming people in China: a systematic review.

Transgender and gender non-conforming (TGNC) individuals are at a high risk of adverse mental health outcomes due to minority stress-the stress faced by individuals categorised as stigmatised social minority groups. This systematic review sought to summarise the key mental health findings of the research on TGNC individuals in mainland China. We also aimed to consolidate research on the topic, identify specific mental health disparities, and offer new perspectives for future research to inform both policy and clinical practice. An extensive search of the literature, published in English and Chinese, was done between Jan 1, 1990, and Aug 1, 2021, using PubMed, PsycINFO, Scopus, Wanfang (in Chinese), and CNKI (in Chinese). Overall, two qualitative and 28 quantitative articles were identified. The quantitative findings showed a high prevalence of mental health problems, such as depression, anxiety, substance use disorders, and stress-related issues, and greater disparities in psychological wellbeing. High prevalence is also reported in suicidality and self-harm behaviours in this group. Across the two qualitative studies, attributable factors included gender-related discrimination, barriers to accessing health services, low social support, decreased knowledge and awareness of HIV prevention, and demographic characteristics-such as marital status, educational level, and gender identity. This Review also found little evidence of gender-affirming care and mental health interventions in mainland China. Following from these results, the next step is to integrate multi-level, social-psychological interventions with education to reduce cultural stereotypes and transphobia in mainland China. Political and social implications are also discussed to inform a standard set of guidelines for transgender-inclusive health-care services, including advocating for funding to create these special care programmes and services.