Public Datasets Research Articles

The Fourth S-PLUS Data Release: 12-filter photometry covering ~3000 square degrees in the southern hemisphere

The Southern Photometric Local Universe Survey (S-PLUS) is a project to map $ sq deg of the sky using twelve bands (seven narrow and five broadbands). Observations are performed with the T80-South telescope, a robotic telescope located at the Cerro Tololo Observatory in Chile. The survey footprint consists of several large contiguous areas, including fields at high and low galactic latitudes, and towards the Magellanic Clouds. S-PLUS uses fixed exposure times to reach point source depths of about $21$ mag in the $griz$ and $20$ mag in the $u$ and the narrow filters. This paper describes the S-PLUS Data Release 4 (DR4), which includes calibrated images and derived catalogues for over 3000 sq deg, covering the aforementioned area. The catalogues provide multi-band photometry performed with the tools DoPHOT and SExtractor -- point spread function ( and aperture photometry, respectively. In addition to the characterization, we also present the scientific potential of the data. We use statistical tools to present and compare the photometry obtained through different methods. Overall we find good agreement between the different methods, with a slight systematic offset of 0.05\,mag between our and aperture photometry. We show that the astrometry accuracy is equivalent to that obtained in previous S-PLUS data releases, even in very crowded fields where photometric extraction is challenging. The depths of main survey (MS) photometry for a minimum signal-to-noise ratio $S/N = 3$ reach from $ for the bluer bands to $ mag on the red. The range of magnitudes over which accurate photometry is obtained is shallower, reaching $ to $ mag depending on the filter. Based on these photometric data, we provide star-galaxy-quasar classification and photometric redshift for millions of objects. We demonstrate the versatility of the data by presenting the results of a project to identify members of four Abell galaxy clusters in the Local Universe. The S-PLUS DR4 data allow for a reliable assessment of cluster membership out to a large radius corresponding to $5 $. The S-PLUS DR4 can be accessed through the survey data portal. All the software used to generate the catalogues for this release and the scientific investigation presented is available in the collaboration GitHub repository. The S-PLUS DR4 consists of a large, calibrated public dataset, providing powerful ways for studying Galactic and extra-galactic objects through an extensive set of (broad and narrow) filters.

MCPL: Multi-model co-guided progressive learning for multimodal aspect-based sentiment analysis

Multimodal Aspect-Based Sentiment Analysis (MABSA) aims to combine image and text information to extract aspect terms from sentences and predict aspect-based sentiment polarities. Previous joint approaches have either viewed this task as collapsed label sequence annotation or modelled each subtask as a unified index sequence generation task, overlooking the explicit relation among its subtasks. This type of method cannot effectively integrate the strengths of task-specific models and adapt to difficult joint prediction tasks. To address this problem, we propose a Multi-model Co-guided Progressive Learning (MCPL) strategy that leverages task-specific models and the correlations among downstream tasks to expand high-quality training datasets, and provides progressive supervision signals to enhance the model’s ability to adapt from simple to difficult tasks. Based on MCPL, we construct a Phase-wise Progressive Learning (PPL) module for MABSA, in which three Task Affinity-based Pseudo-label Generation (TAPG) modules are proposed for task-specific label expansion. The TAPG modules leverage the task correlations among three MABSA downstream tasks for co-guided teaching, and the PPL module utilises their gradual relationship from aspect term extraction to perform sentiment classification and then joint prediction for progressive learning. Experiments on widely used public datasets demonstrate that the proposed MCPL method achieves excellent performance on the three subtasks of MABSA and outperforms most state-of-the-art methods. The source code is publicly available at https://github.com/qujiaqi-babu/MCPL.

Mast cell infiltration and subtype promote malignant transformation of oral precancer and progression of oral cancer.

The role of mast cell (MC), a common myeloid-derived immune cell, in the development of oral squamous cell carcinoma (OSCC) is unclear and the aim of this study was to investigate MC infiltration in oral precancer and oral cancer. Evaluation of immune cell infiltration and association with prognosis in OSCC using RNA sequencing and multiple public datasets. Multiplex immunofluorescence was used to explore the infiltration of MC in the microenvironment of OSCC and oral precancer and the interaction with CD8+ cells. The role of MC in OSCC progression was verified by in vivo experiments. The resting MC infiltration was mainly present in oral precancer, while activated MC infiltration was significantly higher in OSCC. Activated MC was associated with malignant transformation of oral precancer and poor prognosis of OSCC. In vivo studies showed that MC promoted the growth of OSCC. The infiltration of activated MC was negatively correlated with the infiltration of CD8+ T cells. The subtype of MC containing tryptase without chymase (MCT) was significantly higher in OSCC compared to oral precancer and was associated with poor survival. Furthermore, spatial distance analysis revealed a greater distance between MCT and CD8+ cells that was also linked to poor prognosis in OSCC. Cox regression analysis showed that MCT could be a potential diagnostic and prognostic biomarker. This study provides new insights into the role of MC in the immune microenvironment of OSCC. It might enhance the immunotherapeutic efficacy of OSCC through developing targeted therapies against MC.

Proteomic insights into the regulatory function of ARID1A in colon cancer cells.

The AT-rich interacting domain-containing protein 1A (ARID1A) is a tumor suppressor gene that has been implicated in several cancers, including colorectal cancer (CRC). The present study used a proteomic approach to elucidate the molecular mechanisms of ARID1A in CRC carcinogenesis. Stable ARID1A-overexpressing SW48 colon cancer cells were established using lentivirus transduction and the successful overexpression of ARID1A was confirmed by western blotting. Label-free quantitative proteomic analysis using liquid chromatography-tandem mass spectrometry identified 705 differentially altered proteins in the ARID1A-overexpressing cells, with 310 proteins significantly increased and 395 significantly decreased compared with empty vector control cells. Gene Ontology enrichment analysis highlighted the involvement of the altered proteins mainly in the Wnt signaling pathway. Western blotting supported these findings, as a decreased protein expression of Wnt target genes, including c-Myc, transcription factor T cell factor-1/7 and cyclin D1, were observed in ARID1A-overexpressing cells. Among the altered proteins involved in the Wnt signaling pathway, the interaction network analysis revealed that ARID1A exhibited a direct interaction with E3 ubiquitin-protein ligase zinc and ring finger 3 (ZNRF3), a negative regulator of the Wnt signaling pathway. Further analyses using the The Cancer Genome Atlas colon adenocarcinoma public dataset revealed that ZNRF3 expression significantly impacted the overall survival of patients with CRC and was positively correlated with ARID1A expression. Finally, an increased level of ZNRF3 in ARID1A-overexpressing cells was confirmed by western blotting. In conclusion, the findings of the present study suggest that ARID1A negatively regulates the Wnt signaling pathway through ZNRF3, which may contribute to CRC carcinogenesis.

Variation-aware directed graph convolutional networks for skeleton-based action recognition

Directed Graph convolutional networks (DGCNs) have been indeed gaining attention and being applied in skeleton-based action recognition tasks to capture the hierarchical relationships of skeleton via directed graph topology. However, they typically pay the same attention to regions with variations and regions with relative statics of the skeleton, which leads to low accuracy when recognizing fine-grained actions. To this end, we design an innovative variation-aware directed graph convolutional network (VA-DGCN) to focus on the regions where the variation takes place. VA-DGCN comprises a variation-aware directed spatial convolution (VDSC) module and a multi-scale contrastive temporal convolution (MCTC) module. Specifically, in order to capture subtle variations in fine-grained actions, VDSC introduce the average posture of the action sequence as the static anchor. In VDSC, the channel-specific topology branch is designed to model the different kinematic properties of different channels to extract global features, to which global-attention graph convolution is added for unnaturally connected joints. Subsequently, we identify the regions with variations by comparing global features acquired from action sequence and average posture. In temporal, MCTC comprises multiple branches for extracting temporal features at different scales. Moreover, to maximize the mutual information between branches, we introduce contrastive learning to drive the module to learn more meaningful action representations. We conduct extensive experiments on three public datasets to validate the feasibility and efficacy of our proposed VA-DGCN.

CfExplainer: Explainable Just-In-Time Defect Prediction Based on Counterfactuals

Just-in-time (JIT) defect prediction helps rationally allocate testing resources and reduce testing costs. However, most JIT defect prediction models lack explainability, which significantly affects their credibility. Recently, the local interpretable model-agnostic explanations (LIME) method has been used widely in model-explainable research, and many improved LIME-based methods have been proposed. However, problems with respect to explanation effectiveness and reliability remain, which seriously affects the practical use of LIME. To address this problem, CfExplainer, a local rule-based model-agnostic approach, is proposed. The approach first applies counterfactuals to generate synthetic instances. It then mines weighted class association rules based on synthetic instances, and it optimises the process of generating, ranking, pruning, and predicting the class association rules. Next, it employs the rules with the highest priority to explain the prediction results of the model. Experiments were conducted using the public datasets employed in related studies. Compared to other state-of-the-art methods, in terms of explanatory effectiveness, CfExplainer's instance similarity improves by 26.5%-31.2%, and local model fittness improves by 2.0%-3.5%, 2.3%-3%, and 0.7%-7.5% on the AUC, F1-score, and Popt metrics, respectively. In terms of the reliability of the explanation, explanations that are 2.6%-4.7% more unique and 2.5%-5.9% more consistent with the actual characteristics of defect-introducing commits than other state-of-the-art methods. Thus, the explanations of the proposed approach can enhance the model credibility and help guide developers in fixing defects and reducing the risk of introducing them.

M2DCapsN: Multimodal, Multichannel, and Dual-Step Capsule Network for Natural Language Moment Localization.

Natural language moment localization aims to localize the target moment that matches a given natural language query in an untrimmed video. The key to this challenging task is to capture fine-grained video-language correlations to establish the alignment between the query and target moment. Most existing works establish a single-pass interaction schema to capture correlations between queries and moments. Considering the complex feature space of lengthy video and diverse information between frames, the weight distribution of information interaction flow is prone to dispersion or misalignment, which leads to redundant information flow affecting the final prediction. We address this issue by proposing a capsule-based approach to model the query-video interactions, termed the Multimodal, Multichannel, and Dual-step Capsule Network ( [Formula: see text]DCapsN), which is derived from the intuition that "multiple people viewing multiple times is better than one person viewing one time." First, we introduce a multimodal capsule network, replacing the single-pass interaction schema of "one person viewing one time" with the iterative interaction schema of "one person viewing multiple times," which cyclically updates cross-modal interactions and modifies potential redundant interactions via its routing-by-agreement. Then, considering that the conventional routing mechanism only learns a single iterative interaction schema, we further propose a multichannel dynamic routing mechanism to learn multiple iterative interaction schemas, where each channel performs independent routing iteration to collectively capture cross-modal correlations from multiple subspaces, that is, "multiple people viewing." Moreover, we design a dual-step capsule network structure based on the multimodal, multichannel capsule network, bringing together the query and query-guided key moments to jointly enhance the original video, so as to select the target moments according to the enhanced part. Experimental results on three public datasets demonstrate the superiority of our approach in comparison with state-of-the-art methods, and comprehensive ablation and visualization analysis validate the effectiveness of each component of the proposed model.

COVID-19 Mental Health Impact Analysis using Ensemble-based Classifier

Introduction In the 21st century, human community witnessed a range of biological crises resulting in long-term consequences like loss of life, economic decline, trauma and social disruptions. COVID -19, named the SARs-CoV-2 virus by United Nations, was a similar outbreak in China in the year 2019, which later spread across the world. During the pandemic, as part of preventive measures, the government authorities introduced SOP (standard operating procedures) measures such as social distancing, lockdown, quarantining and closure of educational institutions imposing a great impact on mental health and well-being of humans, especially among the youth. Materials and Methods A study was performed on a public dataset containing survey records collected from 1182 students of different educational institutions. The survey data was based on age, region of residence, time spent online and health fitness. The method used in the proposed work is a classifier model based on an ensemble of decision trees called random forest to predict the consequences of online learning on student’s health. The optimum and promising features are selected by using Recursive feature elimination (RFE) method. Results Our findings reveal a notable enhancement in predicting human health during a pandemic, as indicated by a significant increase in validation accuracy based on confusion for various classifiers. Experimental validation of the developed classifier model is done through the confusion matrix and receiver operating characteristic (ROC) curve. Further, performance metrics such as accuracy, precision, recall, F1-score, specificity, and error rate were employed. The experimental results established the superiority of the proposed ensemble subspace discriminant classifier compared to traditional classifiers. Discussions The RFE feature selection method used in the proposed work helps to select the optimum features as well as more informative features. Moreover, the method employed hyper parameter tuning method to enhance the performance of the classifier model. Conclusion This study highlights the importance of taking care of the emotional and physical health of humans during any pandemic. Furthermore, our approach possesses the capacity to significantly influence the field of predicting health, facilitating the development of more effective and advanced prediction strategies in the future.

Joint extraction of Chinese medical entities and relations based on RoBERTa and single-module global pointer

BackgroundMost Chinese joint entity and relation extraction tasks in medicine involve numerous nested entities, overlapping relations, and other challenging extraction issues. In response to these problems, some traditional methods decompose the joint extraction task into multiple steps or multiple modules, resulting in local dependency in the meantime.MethodsTo alleviate this issue, we propose a joint extraction model of Chinese medical entities and relations based on RoBERTa and single-module global pointer, namely RSGP, which formulates joint extraction as a global pointer linking problem. Considering the uniqueness of Chinese language structure, we introduce the RoBERTa-wwm pre-trained language model at the encoding layer to obtain a better embedding representation. Then, we represent the input sentence as a third-order tensor and score each position in the tensor to prepare for the subsequent process of decoding the triples. In the end, we design a novel single-module global pointer decoding approach to alleviate the generation of redundant information. Specifically, we analyze the decoding process of single character entities individually, improving the time and space performance of RSGP to some extent.ResultsIn order to verify the effectiveness of our model in extracting Chinese medical entities and relations, we carry out the experiments on the public dataset, CMeIE. Experimental results show that RSGP performs significantly better on the joint extraction of Chinese medical entities and relations, and achieves state-of-the-art results compared with baseline models.ConclusionThe proposed RSGP can effectively extract entities and relations from Chinese medical texts and help to realize the structure of Chinese medical texts, so as to provide high-quality data support for the construction of Chinese medical knowledge graphs.

Quantum Nearest Neighbor Collaborative Filtering Algorithm for Recommendation System

Recommendation has become especially crucial during the COVID-19 pandemic as a significant number of people rely on online shopping from home. Existing recommendation algorithms, designed to address issues like cold start and data sparsity, often overlook the time constraints of users. Specifically, users expect to receive recommendations for products of interest in the shortest possible time. To address this challenge, we propose a novel collaborative filtering recommendation algorithm that leverages the advantages of quantum computing circuits based on data reconstruction. This approach allows for the rapid identification of users similar to the target user, thereby improving recommendation speed. In our method, we utilize the information of known users to linearly reconstruct that of the target users, forming a relational matrix. Subsequently, we employ \(l_{2,1}-\) norm and \(l_{1}-\) norm to sparsely constrain the relationship matrix, deducing the weight of each known user. The final step involves providing similar recommendations to target users based on these weights. Furthermore, we implement the proposed algorithm using a quantum circuit, enabling exponential acceleration. The final weight matrix is derived from the quantum state outputted by the circuit. The speed of this process is theoretically demonstrated in detail. Experimental results indicate that our algorithm outperforms state-of-the-art methods in terms of root mean squared error (RMSE), mean absolute error (MAE) and normalized discounted cumulative gain (NDCG). Compared to state-of-the-art comparison algorithms, the proposed algorithm achieves the fastest recommendation speed across eight public datasets.

Research on Student’s T-Distribution Point Cloud Registration Algorithm Based on Local Features

LiDAR offers a wide range of uses in autonomous driving, remote sensing, urban planning, and other areas. The laser 3D point cloud acquired by LiDAR typically encounters issues during registration, including laser speckle noise, Gaussian noise, data loss, and data disorder. This work suggests a novel Student’s t-distribution point cloud registration algorithm based on the local features of point clouds to address these issues. The approach uses Student’s t-distribution mixture model (SMM) to generate the probability distribution of point cloud registration, which can accurately describe the data distribution, in order to tackle the problem of the missing laser 3D point cloud data and data disorder. Owing to the disparity in the point cloud registration task, a full-rank covariance matrix is built based on the local features of the point cloud during the objective function design process. The combined penalty of point-to-point and point-to-plane distance is then added to the objective function adaptively. Simultaneously, by analyzing the imaging characteristics of LiDAR, according to the influence of the laser waveform and detector on the LiDAR imaging, the composite weight coefficient is added to improve the pertinence of the algorithm. Based on the public dataset and the laser 3D point cloud dataset acquired in the laboratory, the experimental findings demonstrate that the proposed algorithm has high practicability and dependability and outperforms the five comparison algorithms in terms of accuracy and robustness.

Learning a Memory-Enhanced Multi-Stage Goal-Driven Network for Egocentric Trajectory Prediction

We propose a memory-enhanced multi-stage goal-driven network (ME-MGNet) for egocentric trajectory prediction in dynamic scenes. Our key idea is to build a scene layout memory inspired by human perception in order to transfer knowledge from prior experiences to the current scenario in a top-down manner. Specifically, given a test scene, we first perform scene-level matching based on our scene layout memory to retrieve trajectories from visually similar scenes in the training data. This is followed by trajectory-level matching and memory filtering to obtain a set of goal features. In addition, a multi-stage goal generator takes these goal features and uses a backward decoder to produce several stage goals. Finally, we integrate the above steps into a conditional autoencoder and a forward decoder to produce trajectory prediction results. Experiments on three public datasets, JAAD, PIE, and KITTI, and a new egocentric trajectory prediction dataset, Fuzhou DashCam (FZDC), validate the efficacy of the proposed method.

Heterogeneous Meta-Path Graph Learning for Higher-Order Social Recommendation

Recommendation systems have become an indispensable part of daily life. Social recommendation systems, which utilize social relationships and past behaviors to infer users’ preferences, have gained popularity in recent years. Exploring the inherent characteristics implied by higher-order relationships offers a new approach to social recommendation. However, it is challenging due to sparse social networks, influence heterogeneity, and noisy feedback. In this article, we propose a Heterogeneous Meta-path Graph Learning model for Higher-order Social Recommendation (HEAL). Within HEAL, we introduce a heterogeneous graph in social recommendation and utilize a meta-path-guided random walk to generate higher-order relationships. By encoding higher-order structures and semantics along different meta-graphs, HEAL can mitigate the limitation of data sparsity. Moreover, HEAL exploits aspect-aware and semantic-aware attentions to adaptively propagate and aggregate useful features from different meta-neighbors and higher-order relations. These attention-based aggregation layers allow HEAL to suppress the heterogeneity of social influences. Furthermore, HEAL adopts contrastive learning as a supplemental task to the recommendation task by maximizing the consistency between the self-discriminating objectives. This auxiliary task enables the model to learn more differentiated representations, further reducing its sensitivity to noisy feedback. We evaluate the performance of HEAL through extensive experiments on public datasets. The results demonstrate that leveraging higher-order relations can enhance the quality of social recommendations by better capturing the complexity and diversity of users’ preferences and interactions.

NetraDeep: An Integrated Deep Learning and Image Processing System for Precise Detection of Hard Exudates

Hard Exudate (HE) is a common manifestation of various eye diseases, such as diabetic retinopathy (DR), and a prominent cause of vision loss and blindness. Researchers aim to visualize and quantify these exudates using deep learning (DL) and image processing (IP) models from retinal images. However, the requirement for a large number of labelled image datasets for DL models to work on diverse and poor-quality images makes this task challenging. To address this challenge, we introduce NetraDeep, a system that integrates data-driven DL and rule-based IP techniques for exudate segmentation. Our system uses IP models to detect and extract some features and assists DL models in detecting more advanced features and vice versa. The IP models are rule-based and use predefined rules to process images, while the DL models are data-driven and learn from the input data. NetraDeep provides visual and quantitative assessments while mitigating noise and other confounding factors such as artifacts and noise. The training of DL models of this system requires only a limited number of labelled fundus images from publicly available datasets. It provides accurate pixel-wise segmentation results on the public and private image datasets collected from local eye hospitals. Through extensive evaluation, our system achieved remarkable performance, with a dice coefficient of \(0.84\) for the public dataset and a rating of \(9.78\) and \(9.43\) out of 10, as corroborated by two medical experts with experience of more than 20 and 5 years, respectively, for the private image dataset.

The advancements and future prospects of pedestrian action recognition based on machine learning algorithms

This paper highlights the shift from classical machine learning to deep learning models and describes the fundamental methodology and developments in pedestrian activity recognition. The four main steps of the workflow are the gathering of datasets, pre-processing, designing and training the model, and evaluating the outcome. To extract pedestrian feature vectors, data must first be collected, cleaned, and processed from public or proprietary datasets. These vectors are used to train deep learning or machine learning models, which are subsequently assessed and fine-tuned for use in practical applications such as behaviour analysis and surveillance. For action recognition, conventional machine learning techniques like Random Forests (RF) and Support Vector Machines (SVM) have been used. SVMs, despite their potential for computing complexity, identify the best hyperplanes for classification. The categorization rates for a variety of human behaviours have been enhanced by a combination strategy utilizing SVMs and decision trees. As shown in a study that uses smartphone accelerometers to accurately identify everyday activities, RFs can manage enormous datasets. Deep learning models that automatically learn complicated feature representations, such as VA-fusion, AGC-LSTM, and LC-POSEGAIT, provide improved performance. These models capture minute differences in pedestrian behaviour using CNNs, RNNs, and LSTM architectures. Interpretability, generalization to new datasets, and computing demands are some of the difficulties they encounter. Future developments could include using transfer learning to improve performance in many circumstances, combining deep learning and expert systems for improved interpretability, and utilizing distributed computing for processing in an effective manner.

Integrative mRNA and miRNA Expression Profiles from Developing Zebrafish Head Highlight Brain-Preference Genes and Regulatory Networks.

Zebrafish is an emerging animal model for studying molecular mechanism underlying neurodevelopmental disorder due to its advantage characters. miRNAs are small non-coding RNAs that play a key role in brain development. Understanding of dynamic transcriptional and post-transcriptional molecules and their regulation during the head development is important for the study of neurodevelopmental disorder. In this study, we performed the high-throughput sequencing of mRNAs and miRNAs in developing zebrafish head from pharyngula to early larval stages and carried out bioinformatic analysis including differential expression and functional enrichment as well as joint analysis of miRNAs and mRNAs, and also compared with other related public sequencing datasets to aid our interpretation. A large number of differential expression genes with a large fold change were detected during the head development. Further clustering and functional enrichment analyses indicated that genes in late stage were most related with synaptic signaling. Overlap test analysis showed a significant enrichment of brain-preference and synapse-associated gene set in the head transcriptome compared with the whole embryo transcriptome. We also constructed miRNA-mRNA network for those brain-preference genes and focused on those densely connected network components. CRISPR-Cas9-mediated snap25b mutants led to embryonic development defects and decreases locomotor activity. Altogether, the present study provides developmental profiles of head-enriched mRNAs and miRNAs at three critical windows for nervous system development, which may contribute to the study of neurodevelopmental disorder.

A Practical Approach to Intrusion Detection in IoV Networks Using LCCDE Ensemble Framework

Abstract: With the rapid development of the Internet of Things (IoT) and the Internet of Vehicles (IoV) technologies, modern vehicles are increasingly adopting network-controlled functionalities, exposing them to a variety of cyber threats. To address these security challenges, this paper implements and evaluates a novel ensemble Intrusion Detection System (IDS) framework, named Leader Class and Confidence Decision Ensemble (LCCDE). The LCCDE framework integrates three advanced Machine Learning (ML) algorithms—XGBoost, LightGBM, and CatBoost—to detect various types of cyber-attacks on both intra-vehicle and external vehicular networks.This study demonstrates the effectiveness of the LCCDE framework using two public IoV security datasets: the Car-Hacking dataset and the CICIDS2017 dataset. By identifying the best-performing ML model for each class of attacks and leveraging prediction confidence values, LCCDE achieves high detection accuracy and robustness against diverse attack types. The experimental results show that the proposed framework outperforms traditional IDS approaches in terms of detection accuracy, computational efficiency, and adaptability to different types of cyber-attacks. This paper provides practical insights into the implementation and deployment of IDS in IoV systems, highlighting the potential of ensemble learning methods to enhance vehicular network security.

Semi-Supervised Instance Segmentation in Whole Slide Images via Dense Spatial Variability Enhancing.

Current whole slide image (WSI) segmentation aims at extracting tumor regions from the background. Unlike this, segmenting distinct tumor areas (instances) within a WSI driven by limited annotated data remains under-explored. In this paper, we formally propose semisupervised instance segmentation (Semi-IS) in WSIs. We address a key challenge: learning intra-class similarity and inter-class dissimilarity driven by unlabeled data. Specifically, we generally perceive the patch as composed of tokens (together), not the patch alone. We employ contrastive learning to develop a segmentation framework. In the SemiIS, we find that the boundaries of segmented instances are usually disturbed by noise. We jointly eliminate and preserve noise features to address this problem. We conduct extensive experiments to evaluate the effectiveness and generalizability of Semi-IS, including histopathology and cellular pathology. The results show that in clinical multi instance segmentation tasks, Semi-IS achieves almost fullsupervised state-of-the-art results with only 30% annotated data. Semi-IS can improve segmentation accuracy by about 2% on public cell pathology datasets.

NCNet: deformable medical image registration network based on neighborhood cross-attention combined with multi-resolution constraints.


Existing registration networks based on cross-attention design usually divide the image pairs to be registered into patches for input. The division and merging operations of a series of patches are difficult to maintain the topology of the deformation field and reduce the interpretability of the network. Therefore, our goal is to develop a new network architecture based on a cross-attention mechanism combined with a multi-resolution strategy to improve the accuracy and interpretability of medical image registration.
Approach: We propose a new deformable image registration network NCNet based on neighborhood cross-attention combined with multi-resolution strategy. The network structure mainly consists of a multi-resolution feature encoder, a multi-head neighborhood cross-attention module and a registration decoder. The hierarchical feature extraction capability of our encoder is improved by introducing large kernel parallel convolution blocks; the cross-attention module based on neighborhood calculation is used to reduce the impact on the topology of the deformation field and double normalization is used to reduce its computational complexity.
Main result: We performed atlas-based registration and inter-subject registration tasks on the public 3D brain magnetic resonance imaging datasets LPBA40 and IXI respectively. Compared with the popular VoxelMorph method, our method improves the average DSC value by 7.9% and 3.6% on LPBA40 and IXI. Compared with the popular TransMorph method, our method improves the average DSC value by 4.9% and 1.3% on LPBA40 and IXI.
Significance: We proved the advantages of the neighborhood attention calculation method compared to the window attention calculation method based on partitioning patches, and analyzed the impact of the pyramid feature encoder and double normalization on network performance. This has made a valuable contribution to promoting the further development of medical image registration methods.&#xD.

Modeling QTL-by-environment interactions for multi-parent populations

Multi-parent populations (MPPs) are attractive for genetic and breeding studies because they combine genetic diversity with an easy-to-control population structure. Most methods for mapping QTLs in MPPs focus on the detection of QTLs in single environments. Little attention has been given to mapping QTLs in multienvironment trials (METs) and to detecting and modeling QTL-by-environment interactions (QEIs). We present mixed model approaches for the detection and modeling of consistent versus environment-dependent QTLs, i.e., QTL-by-environment interaction (QEI). QTL effects are assumed to be normally distributed with variances expressing consistency or dependence on environments and families. The entries of the corresponding design matrices are functions of identity-by-descent (IBD) probabilities between parents and offspring and follow from the parental origin of offspring DNA. A polygenic effect is added to the models to account for background genetic variation. We illustrate the wide applicability of our method by analyzing several public MPP datasets with observations from METs. The examples include diallel, nested association mapping (NAM), and multi-parent advanced inter-cross (MAGIC) populations. The results of our approach compare favorably with those of previous studies that used tailored methods.