Scarcity Of Data Research Articles

Sparse cross-transformer network for surface defect detection.

Quality control processes with automation ensure that customers receive defect-free products that meet their needs. However, the performance of real-world surface defect detection is often severely hindered by the scarcity of data. Recently, few-shot learning has been widely proposed as a solution to the data sufficiency problem by leveraging a limited number of base class samples. However, achieving discriminative and generalization capabilities with few samples remains a challenging task in various surface defect detection scenarios. In this paper, we propose a sparse cross-transformer network (SCTN) for surface defect detection. Specifically, we introduce a residual layer module to enhance the network's ability to retain crucial information. Next, we propose a sparse layer module within the cross-transformer to increase computational efficiency. Finally, we incorporate a squeeze-and-excitation network into the cross-transformer to enhance the attention mechanism between local patches outputted by the transformer encoder. To verify the effectiveness of our proposed method, we conducted extensive experiments on the cylinder liner defect dataset, the NEU steel surface defect dataset, and the PKU-Market-PCB dataset, achieving the best mean average precision of 62.73%, 85.29%, and 88.7%, respectively. The experimental results demonstrate that our proposed method achieves significant improvements compared to state-of-the-art algorithms. Additionally, the results indicate that SCTN enhances the network's discriminative ability and effectively improves generalization across various surface defect detection tasks.

The Ways of Forming and the Erosion/Decay/Aging of Bioapatites in the Context of the Reversibility of Apatites.

This study primarily focused on the acid erosion of enamel and dentin. A detailed examination of the X-ray diffraction data proves that the products of the acid-caused decay of enamel belong to the family of isomorphic bioapatites, especially calcium-deficient hydroxyapatites. They are on a trajectory towards less and less crystallized substances. The increase in Bragg's parameter d and the decrease in the energy necessary for the changes were coupled with variability in the pH. This was valid for the corrosive action of acid solutions with a pH greater than 3.5. When the processes of natural tooth aging were studied by X-ray diffraction, a clear similarity to the processes of the erosion of teeth was revealed. Scarce data on osteoporotic bones seemed to confirm the conclusions derived for teeth. The data concerning the bioapatite decays were confronted with the cycles of apatite synthesis/decay. The chemical studies, mainly concerning the Ca/P ratio in relation to the pH range of durability of popular compounds engaged in the synthesis/decay of apatites, suggested that the process of the formation of erosion under the influence of acids was much inverted in relation to the process of the formation of apatites, starting from brushite up to apatite, in an alkaline environment. Our simulations showed the shift between the family of bioapatites versus the family of apatites concerning the pH of the reaction environment. The detailed model stoichiometric equations associated with the particular stages of relevant processes were derived. The synthesis processes were alkalization reactions coupled with dehydration. The erosion processes were acid hydrolysis reactions. Formally, the alkalization of the environment during apatite synthesis is presented by introducing Ca(OH)2 to stoichiometric equations.

Vuv/vis absorption spectroscopy of different PAHs

AbstractThe present study provides new solid-phase absorption data of different Polycyclic Aromatic Hydrocarbon (PAHs) molecules in the Far Ultraviolet (FUV) and UV/Visible regions. Two features of interstellar extinction curves fall in this region: the FUV rise and the UV bump. Here, thin films of PAHs were prepared by means of the spin coating method onto a transparent substrate, and their solid phase absorption spectra were recorded. Spectra of isolated molecules simulated by Time-Dependent Density Functional Theory (TD-DFT) calculations were also produced for comparison. The results fit quite well with the FUV rise but not with the UV bump. Given the scarcity of FUV experimental absorption data of PAHs in the literature and especially in solid phase, these results are also interesting for reference for other studies ranging from thin films applications to interstellar medium models.

Assessing long-term water storage dynamics in Afghanistan: An integrated approach using machine learning, hydrological models, and remote sensing

Assessment of Terrestrial Water Storage (TWS) components is crucial for understanding regional climate and water resources, particularly in arid and semi-arid regions like Afghanistan. Given the scarcity of ground-based data, this study leverages remote sensing datasets to quantify water storage changes. We integrated Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-on (GRACE-FO) data with WaterGap, Global Land Water Storage (GWLS), Catchment Land Surface Model (CLSM), and climate variables (precipitation, temperature, potential evapotranspiration) using artificial neural networks (ANN) and random forests (RF). Additionally, Ice, Cloud, and Land Elevation Satellite (ICESat-1,2) data were utilized to estimate glacier mass changes. Seasonal trend decomposition using Loess (STL) was applied to assess TWS changes from 2003 to 2022. Our methodology reveals a high correlation (R = 0.90–0.97) between reconstructed and observed TWS anomalies across Afghanistan's major basins. Glacier mass decreased by −0.59 and −1.17 Gt/year during 2003–2009 and 2018–2022, respectively, while overall TWS declined by −2.46 Gt/year. The HRB experienced the largest TWS loss (−1.47 Gt/year), primarily due to groundwater depletion (−1.18 Gt/year). These findings underscore the importance of our approach for assessing water resources, providing vital insights for sustainable management under a changing climate in a data-scarce country.

Deep learning-powered efficient characterization and quantification of microplastics

Characterizing and quantifying microplastics (MPs) are time-consuming and labor-intensive tasks traditionally. This paper presents an artificial intelligence (AI) framework aiming to automate these tasks by integrating computer vision and deep learning techniques. The approach leverages Fourier Transform Infrared (FTIR) spectra and visual images. Primary novelties of this research involve the development of: (1) an AI framework integrating efforts of data processing, analytics, visualization, and human-computer interaction; (2) a method for transforming FTIR data into contour images; (3) data augmentation strategies for resolving data scarcity and imbalance issues; (4) deep learning models for identifying MPs; (5) computer vision algorithms for quantifying MPs; and (6) an engineer-friendly graphic user interface (GUI) for enhancing data accessibility. The AI framework has been applied to polyethylene, polypropylene, polystyrene, polyamide, ethylene-vinyl acetate, and cellulose acetate. Results confirmed the efficacy of the framework, exhibiting high accuracy scores in classification (98 %), segmentation (99 %), and quantification (96 %) tasks. This research advances the capability of automatic assessment of MPs.

The landscape of city‐level GHG emission accounts in Africa

AbstractCities are important actors in the global challenge of tackling climate change. They are not only responsible for the majority of emissions but also highly capable of taking action. An important precondition for effective climate mitigation is a city‐level greenhouse gas emission inventory to guide mitigation action. Yet, most cities in developing countries, in particular African cities, lack that crucial information. This study aims to assess the current state of the development of African cities' emission inventories. A total of 270 inventories from 137 cities across 54 African countries were identified from 15 research articles, 5 reports, and 3 data platforms. We find the lack of standardized protocols results in inventories that are often not comparable, while data scarcity emerged as a common problem. We observe that insufficient engagement from local governments impedes the creation of a data‐rich environment. Additionally, current inventory protocols do not fully address the data limitations faced by African cities, further hindering inventory development. To mitigate these challenges, multi‐agent collaboration is essential to enhance the accounting capabilities of local governments. Developing refined protocols that consider data constraints is necessary. Moreover, advanced technologies may provide opportunities to overcome data bottlenecks.

Immunological resilience of a temperate catshark to a simulated marine heat wave.

Marine heatwaves (MHW) have recently been proposed as more relevant in driving population changes than the continuous increase in average temperatures associated with climate change. The causal processes underpinning MHW effects in sharks are unclear but may be linked to changes in fitness caused by physiological trade-offs that influence the immune response. Considering the scarcity of data about the immune response of sharks under anomalous warming events, the present study analyzed several fitness indices and characterized the immune response (in the blood, epigonal organ, liver, spleen, and intestine) of temperate adult small-spotted catsharks (Scyliorhinus canicula) after a 30-day exposure to a Category II MHW. The results indicated that adult small-spotted catsharks have developed coping strategies for the MHW. Specifically, among the 35 parameters investigated, only the gonad-to-body ratio (GBR) and plasma glucose showed significant increases. In contrast, igm and tumor necrosis factor receptor (tnfr) gene expression in blood cells, tnfr in the epigonal organ, and the number of monocytes significantly decreased. Although a decline in immune function in small-spotted catsharks was revealed following the MHW exposure, energy mobilization restored homeostasis and indicated a shift in energy allocation towards reproduction. Group resilience may be due to the variable tolerance of individuals, the phenotypic plasticity of cellular immunity, thermal imprinting, and/or metabolic capacity of the individuals.

Genome-wide exploration of the CONSTANS-like (COL) gene family and its potential role in regulating plant flowering time in foxtail millet (Setaria italica)

In photoperiod-sensitive plants, the CO-like gene (CONSTANS-like, COL) has a crucial function in regulating the timing of flowering. The blooming period in foxtail millet is greatly influenced by the duration of daylight; however, there is a scarcity of data regarding the molecular properties of the COL genes in the foxtail millet. In this study, we conducted a comprehensive analysis of the COL gene family in foxtail millet at the genome-wide level. We identified 11 SiCOL genes and performed gene structure analysis, which showed pronounced variation in gene length and intron number among the genes. The examination of COL proteins in foxtail millet and other plant species using phylogenetic analysis revealed that they could be clustered into three distinct groups. Cis-element analysis identified elements related to light-responsiveness, hormones, and abiotic stress in the promoter region of the SiCOL gene. Furthermore, tissue-specific expression analysis showed widespread expression of all 11 SiCOL genes in various foxtail millet tissues and organs, particularly in leaves and panicles. Collinearity analysis identified 14 syntenic gene pairs in both foxtail millet and rice. The results also revealed diurnal oscillations in the transcription levels of SiCOL genes under different light conditions. Moreover, among the 11 genes, SiCO, SiCOL1, and SiCOL6 expression levels were negatively correlated with flowering time variation in two foxtail cultivars. Additionally, upon constructing a network of predicted molecular interactions, FLOWER LOCUS-like (FTL) and Phytochromea A (PHY A) were suggested to potentially interact with SiCO, SiCOL1, and SiCOL6. SiCO, SiCOL1, and SiCOL6 have the potential for flowering and heading in foxtail millet. This research enhances our comprehension of the role and control of the SiCOL gene family constituents in foxtail millet, establishing a basis for future investigations.

Falls and fall-related injuries: prevalence, characteristics, and treatment among participants of the Geelong Osteoporosis Study.

Falls are a significant public health challenge, especially among older adults. In Australia, falls and related injuries incur an annual cost of $2.3 billion. However, there is a scarcity of prevalence data on falls among population-based groups. This study aimed to report the characteristics, circumstances, and treatment for falls and fall-related injuries in a population-based sample of Australian men and women. Participants from the Geelong Osteoporosis Study provided cross-sectional data: baseline for men (2001-2006) and 10-year follow-up for women (2004-2008). Falls over the previous 12 months were self-reported by 2,631 participants aged 20-97 years (1,533 men and 1,098 women). The study described the timing, location, cause, and treatment of falls. Fall prevalence was standardized to the 2006 Australian population. Data collection included self-reported prior fractures, medication, disease conditions, anthropometry, and biochemical tests. Binary logistic regression identified factors linked to fall-related injuries. Fall rates by age group: 20-29 (24.4%), 30-39 (21.5%), 40-49 (18.7%), 50-59 (24.9%), 60-69 (25.0%), 70-79 (34.6%), 80+ (40.5%). The age-standardized prevalence of falls was 25.0% (95% CI: 23.4-26.7%). In adults ≥65 years, the age-standardized prevalence of falls was 32.4% (95% CI: 29.3-35.5%). Fallers were typically older, less mobile, had higher BMI and cholesterol levels, and were more likely to have musculoskeletal conditions, cancer, and polypharmacy. Men had a higher fall risk, but fewer fall-related fractures compared to women. Most falls occurred outdoors (62.1%), were due to extrinsic cause (58.5%), and were on the same level (78.5%). Factors associated with fall-related injuries included being female, elevated falls and falls outside the home. Nearly half (45.7%) of those injured sought medical attention, primarily from general practitioners (25.7%) and emergency departments (12.6%). Falls are common, occurring in approximately one quarter of adults in this study, with a greater prevalence among those ≥65 years. Women experienced fewer multiple falls than men. Many participants sought medical attention, primarily from general practitioners. This research identifies fall characteristics and circumstances, informing targeted prevention strategies to reduce occurrences and alleviate burdens on healthcare systems and individuals.

Multimodal Seed Data Augmentation for Low-Resource Audio Latin Cuengh Language

Latin Cuengh is a low-resource dialect that is prevalent in select ethnic minority regions in China. This language presents unique challenges for intelligent research and preservation efforts, primarily due to its oral tradition and the limited availability of textual resources. Prior research has sought to bolster intelligent processing capabilities with regard to Latin Cuengh through data augmentation techniques leveraging scarce textual data, with modest success. In this study, we introduce an innovative multimodal seed data augmentation model designed to significantly enhance the intelligent recognition and comprehension of this dialect. After supplementing the pre-trained model with extensive speech data, we fine-tune its performance with a modest corpus of multilingual textual seed data, employing both Latin Cuengh and Chinese texts as bilingual seed data to enrich its multilingual properties. We then refine its parameters through a variety of downstream tasks. The proposed model achieves a commendable performance across both multi-classification and binary classification tasks, with its average accuracy and F1 measure increasing by more than 3%. Moreover, the model’s training efficiency is substantially ameliorated through strategic seed data augmentation. Our research provides insights into the informatization of low-resource languages and contributes to their dissemination and preservation.

Publicly Available Dental Image Datasets for Artificial Intelligence.

The development of artificial intelligence (AI) in dentistry requires large and well-annotated datasets. However, the availability of public dental imaging datasets remains unclear. This study aimed to provide a comprehensive overview of all publicly available dental imaging datasets to address this gap and support AI development. This observational study searched all publicly available dataset resources (academic databases, preprints, and AI challenges), focusing on datasets/articles from 2020 to 2023, with PubMed searches extending back to 2011. We comprehensively searched for dental AI datasets containing images (intraoral photos, scans, radiographs, etc.) using relevant keywords. We included datasets of >50 images obtained from publicly available sources. We extracted dataset characteristics, patient demographics, country of origin, dataset size, ethical clearance, image details, FAIRness metrics, and metadata completeness. We screened 131,028 records and extracted 16 unique dental imaging datasets. The datasets were obtained from Kaggle (18.8%), GitHub, Google, Mendeley, PubMed, Zenodo (each 12.5%), Grand-Challenge, OSF, and arXiv (each 6.25%). The primary focus was tooth segmentation (62.5%) and labeling (56.2%). Panoramic radiography was the most common imaging modality (58.8%). Of the 13 countries, China contributed the most images (2,413). Of the datasets, 75% contained annotations, whereas the methods used to establish labels were often unclear and inconsistent. Only 31.2% of the datasets reported ethical approval, and 56.25% did not specify a license. Most data were obtained from dental clinics (50%). Intraoral radiographs had the highest findability score in the FAIR assessment, whereas cone-beam computed tomography datasets scored the lowest in all categories. These findings revealed a scarcity of publicly available imaging dental data and inconsistent metadata reporting. To promote the development of robust, equitable, and generalizable AI tools for dental diagnostics, treatment, and research, efforts are needed to address data scarcity, increase diversity, mandate metadata completeness, and ensure FAIRness in AI dental imaging research.

Use of Fatty Acids in Fertilizer Formulation: A Systematic Review

Synthetic fertilizers have been a subject of socio-environmental challenges. A more sustainable approach is necessary to develop these farm products. Thus, this review presents a strategy for fertilizer production by utilizing fatty acids and by-products derived from renewable sources. However, given the scarcity of data on the topic in scientific journals, this review used the PRISMA report methodology from patent databases. Results show that China is this field’s most significant intellectual property holder. Patents were predominantly vegetable-based (60%), mixed (14%), and animal- based (8.47%), with innovations including liquid fertilizers of plant and animal-based fatty acid esters, as well as the use of controlled-release technology. 80.60% of fertilizers were applied to the soil, followed by foliar application (12.75%) and seed application (0.60%). Notable contributions included fertilizers acting as pesticides and anti-caking agents. Thus, we realize the urgency of continuing scientific research to pursue more responsible and efficient agricultural practices.

Zero-Shot Relation Classification Through Inference on Category Attributes.

The goal of relationship classification (RC) is to predict the semantic relationship between two entities in a given sentence. With the advent of deep learning and pretrained language models, RC research has progressed by leaps and bounds. However, the current studies are focused mainly on predicting semantic relationships from a predefined set. How to recognize unseen relationships remains a challenge, which is also known as the zero-shot RC (ZSRC) task. Some ZSRC-related methods directly map relationship categories to numerical indices, constraining the model's ability to autonomously infer and understand these relationships, while others rely heavily on manual definitions. To address these issues and inspired by the way of reasoning in which humans perform RC tasks, we propose a new framework to handle the ZSRC task through inference on category attributes (ICAs). The main idea of ICA is to detect the semantic relationship between promises, which are RC sentences, and hypotheses, which are relational sentences of entities created by templates. Specifically, instead of manual design, we introduce two hypothesis templates derived from the label words (LWs) and descriptions (LDs) associated with each relationship. These templates are used to automatically convert the RC data into the textual entailment (TE) format. Furthermore, they are fine-tuned with a pretrained TE model, facilitating the acquisition of relational knowledge and enabling the generalization of semantic reasoning rules learned from seen classes to unseen classes. Moreover, to implement multirelationship semantic inference for all unseen classes, we propose an entailment difference mechanism to enhance the reasoning capability of the model. Besides the current ZSRC test setting, we also examine our method in an even more challenging setting to deal with data scarcity in real-world applications. The outstanding performance of ICA on the FewRel and Wiki-ZSL datasets demonstrates its effectiveness in the ZSRC task.

Strata Design for Variance Reduction in Stochastic Simulation

Stratified sampling is one of the powerful variance reduction methods for analyzing system performance, such as reliability, with stochastic simulation. It divides the input space into disjoint subsets, called strata, to draw samples from each stratum. Partitioning the input space properly and allocating greater computational effort to crucial strata can help accurately estimate system performance with a limited computational budget. How to create strata, however, has yet to be thoroughly examined. Strata design faces the curse of dimensionality and data scarcity as the input dimension increases. We analytically derive the optimal stratification structure that minimizes the estimation variance for univariate problems. Further, reconciling the optimal stratification into decision trees, we devise a robust algorithm for multi-dimensional problems. Numerical experiments and a wind turbine case study demonstrate the superiority of the proposed method in terms of variance reduction, leading to computational efficiency and scalability.

Artificial intelligence for predicting arctic permafrost and active layer temperatures along the Alaskan North Slope

AbstractClimate change in Arctic regions poses a threat to permafrost stability, potentially leading to issues such as infrastructure damage, altered ecosystems, and increased greenhouse gas emissions. The challenge of monitoring permafrost temperatures and identifying critical environmental factors is accentuated by the scarcity of subsurface thermal data in remote Arctic locales. To alleviate this lack of soil thermal data we set out to combine in situ observed soil temperatures with MERRA-2 reanalysis data and Machine Learning (ML) to develop local and season-specific subsurface soil temperature predictions in Alaska. First, reanalysis features are selected based on surface energy budget physics. Coupled with Julian calendar dates, reanalysis features are preprocessed and then fed to the ML prediction model. We conducted a series of computational experiments and tested the results against performance benchmarks to determine the optimal prediction models, length of look-back period of model inputs, and the training set size. Six conventional ML models (e.g., GBDT, RF, SVR) and five statistical baselines (e.g. ARIMA) using in situ time series of soil temperature data ranging from 0 - 1 m depth are considered for the prediction task. The models are trained using in situ soil temperatures at depths between 0 and 1 m, spanning 16+ years, from field sites at Deadhorse and Toolik Lake, Alaska. All prediction performances are assessed using root mean squared error (RMSE) and mean absolute error (MAE). Results show that locally trained ML models can estimate shallow soil temperatures with an average error of $$RMSE=1.308^{\circ }$$ R M S E = 1 . 308 ∘ C.

The biogeochemical model Biome-BGCMuSo v6.2 provides plausible and accurate simulations of the carbon cycle in central European beech forests

Abstract. Process-based ecosystem models are increasingly important for predicting forest dynamics under future environmental conditions, which may encompass non-analogous climate coupled with unprecedented disturbance regimes. However, challenges persist due to the extensive number of model parameters, scarce calibration data, and trade-offs between the local precision and the applicability of the model over a wide range of environmental conditions. In this paper, we describe a protocol that allows a modeller to collect transferable ecosystem properties based on ecosystem characteristic criteria and to compile the parameters that need to be described in the field. We applied the procedure to develop a new parameterisation for European beech (Fagus sylvatica L.) for the Biome-BGCMuSo model, the most advanced member of the Biome-BGC family. For model calibration and testing, we utilised multiyear forest carbon data from 87 plots distributed across five European countries. The initial values of 48 new ecophysiological parameters were defined based on a literature review. The final values of six calibrated parameters were optimised for single sites as well as for multiple sites using generalised likelihood uncertainty estimation (GLUE) and model output conditioning that ensured plausible simulations based on user-defined ranges of carbon stock output variables (carbon stock in aboveground wood biomass, soil, and litter) and finding the intersections of site-specific plausible parameter hyperspaces. To support the model use, we tested the model performance by simulating aboveground tree wood, soil, and litter carbon across a large geographical gradient of central Europe and evaluated the trade-offs between parameters tailored to single plots and parameters estimated using multiple sites. Our findings indicated that parameter sets derived from single sites provided an improved local accuracy of simulations of aboveground wood, soil, and litter carbon stocks by 35 %, 55 %, and 11 % in comparison to the a priori parameter set. However, their broader applicability was very limited. A multi-site optimised parameter set, on the other hand, performed satisfactorily across the entire geographical domain studied here, including on sites not involved in the parameter estimation, but the errors were, on average, 26 %, 35 % and 9 % greater for the aboveground wood, soil, and litter carbon stocks than those obtained with the site-specific parameter sets. Importantly, model simulations demonstrated plausible responses across large-scale environmental gradients, featuring a clear production optimum of beech that aligns with empirical studies. These findings suggest that the model is capable of accurately simulating the dynamics of European beech across its range and can be used for more comprehensive experimentations.

Integrating material flow analysis and supply chain resilience analysis to study silicon carbide

AbstractSilicon carbide (SiC) is a niche nonmetallic material that is essential in many industrial processes. Here, we integrate material flow analysis and supply chain resilience analysis to understand global SiC stocks and flows and to assess its supply chain. We use industry interviews to fill data gaps and collect information on the SiC system to overcome data scarcity. We find that globally around 1000 kt of SiC is produced each year. The biggest use of SiC is the abrasives industry (40%), followed by metallurgy (28%), refractories (20%), technical ceramics (0.7%), other uses (0.7%), and semiconductors (0.01%). As an energy‐intensive material, the SiC supply chain is under pressure, increasing the relevance of resilience considerations. Besides typical supply chain risks such as low diversity of supply and geopolitical trade restrictions, SiC particularly faces risks due to its energy‐intensive production process and associated emissions. In the SiC semiconductor supply chain, losses of nearly 75% are a particular issue. Due to high demand in the SiC market, stockpiles are negligible, and substitution is difficult in most sectors. We find that in the case of SiC, sustainability measures such as use reduction, recycling, or decreasing energy use or emissions would also positively contribute to supply chain resilience. This article met the requirements for a gold‐gold JIE data openness badge described at http://jie.click/badges.

Diffusion probabilistic priors for zero-shot low-dose CT image denoising.

Denoising low-dose computed tomography (CT) images is a critical task in medical image computing. Supervised deep learning-based approaches have made significant advancements in this area in recent years. However, these methods typically require pairs of low-dose and normal-dose CT images for training, which are challenging to obtain in clinical settings. Existing unsupervised deep learning-based methods often require training with a large number of low-dose CT images or rely on specially designed data acquisition processes to obtain training data. To address these limitations, we propose a novel unsupervised method that only utilizes normal-dose CT images during training, enabling zero-shot denoising of low-dose CT images. Our method leverages the diffusion model, a powerful generative model. We begin by training a cascaded unconditional diffusion model capable of generating high-quality normal-dose CT images from low-resolution to high-resolution. The cascaded architecture makes the training of high-resolution diffusion models more feasible. Subsequently, we introduce low-dose CT images into the reverse process of the diffusion model as likelihood, combined with the priors provided by the diffusion model and iteratively solve multiple maximum a posteriori (MAP) problems to achieve denoising. Additionally, we propose methods to adaptively adjust the coefficients that balance the likelihood and prior in MAP estimations, allowing for adaptation to different noise levels in low-dose CT images. We test our method on low-dose CT datasets of different regions with varying dose levels. The results demonstrate that our method outperforms the state-of-the-art unsupervised method and surpasses several supervised deep learning-based methods. Our method achieves PSNR of 45.02 and 35.35dB on the abdomen CT dataset and the chest CT dataset, respectively, surpassing the best unsupervised algorithm Noise2Sim in the comparative methods by 0.39 and 0.85dB, respectively. We propose a novel low-dose CT image denoising method based on diffusion model. Our proposed method only requires normal-dose CT images as training data, greatly alleviating the data scarcity issue faced by most deep learning-based methods. At the same time, as an unsupervised algorithm, our method achieves very good qualitative and quantitative results. The Codes are available in https://github.com/DeepXuan/Dn-Dp.

C2R: A Novel ANN Architecture for Boosting Indoor Positioning With Scarce Data

C2R: A Novel ANN Architecture for Boosting Indoor Positioning With Scarce Data

Detecting m6A RNA modification from nanopore sequencing using a semisupervised learning framework.

Direct nanopore-based RNA sequencing can be used to detect posttranscriptional base modifications, such as N6-methyladenosine (m6A) methylation, based on the electric current signals produced by the distinct chemical structures of modified bases. A key challenge is the scarcity of adequate training data with known methylation modifications. We present Xron, a hybrid encoder-decoder framework that delivers a direct methylation-distinguishing basecaller by training on synthetic RNA data and immunoprecipitation (IP)-based experimental data in two steps. First, we generate data with more diverse modification combinations through in silico cross-linking. Second, we use this data set to train an end-to-end neural network basecaller followed by fine-tuning on IP-based experimental data with label smoothing. The trained neural network basecaller outperforms existing methylation detection methods on both read-level and site-level prediction scores. Xron is a standalone, end-to-end m6A-distinguishing basecaller capable of detecting methylated bases directly from raw sequencing signals, enabling de novo methylome assembly.