Semi-supervised Approach Research Articles

Construction of Multi-Modal Transcriptome-Small Molecule Interaction Networks from High-Throughput Measurements to Study Human Complex Traits.

Small molecules (SMs) are integral to biological processes, influencing metabolism, homeostasis, and regulatory networks. Despite their importance, a significant knowledge gap exists regarding their downstream effects on biological pathways and gene expression, largely due to differences in scale, variability, and noise between untargeted metabolomics and sequencing-based technologies. To address these challenges, we developed a multi-omics framework comprising a machine learning-based protocol for data processing, a semi-supervised network inference approach, and network-guided analysis of complex traits. The ML protocol harmonized metabolomic, lipidomic, and transcriptomic data through batch correction, principal component analysis, and regression-based adjustments, enabling unbiased and effective integration. Building on this, we proposed a semi-supervised method to construct transcriptome-SM interaction networks (TSI-Nets) by selectively integrating SM profiles into gene-level networks using a meta-analytic approach that accounts for scale differences and missing data across omics layers. Benchmarking against three conventional unsupervised methods demonstrated the superiority of our approach in generating diverse, biologically relevant, and robust networks. While single-omics analyses identified 18 significant genes and 3 significant SMs associated with insulin sensitivity (IS), network-guided analysis revealed novel connections between these markers. The top-ranked module highlighted a cross-talk between fiber-degrading gut microbiota and immune regulatory pathways, inferred by the interaction of the protective SM, N-acetylglycine (NAG), with immune genes ( FCER1A , HDC , MS4A2 , and CPA3 ), linked to improved IS and reduced obesity and inflammation. Together, this framework offers a robust and scalable solution for multi-modal network inference and analysis, advancing SM pathway discovery and their implications for human health. Leveraging data from a population of thousands of individuals with extended longevity, the inferred TSI-Nets demonstrate generalizability across diverse conditions and complex traits. These networks are publicly available as a resource for the research community.

Semi-Supervised Medical Image Segmentation Based on Frequency Domain Aware Stable Consistency Regularization.

With the advancement of deep learning models nowadays, they have successfully applied in the semi-supervised medical image segmentation where there are few annotated medical images and a large number of unlabeled ones. A representative approach in this regard is the semi-supervised method based on consistency regularization, which improves model training by imposing consistency constraints (perturbations) on unlabeled data. However, the perturbations in this kind of methods are often artificially designed, which may introduce biases unfavorable to the model learning in the handling of medical image segmentation. On the other hand, the majority of such methods often overlook the supervision in the Encoder stage of training and primarily focus on the outcomes in the later stages, potentially leading to chaotic learning in the initial phase and subsequently impacting the learning process of the model in the later stages. At the meanwhile, they miss the intrinsic spatial-frequency information of the images. Therefore, in this study, we propose a new semi-supervised medical image segmentation approach based on frequency domain aware stable consistency regularization. Specifically, to avoid the bias introduced by artificially setting perturbations, we first utilize the inherent frequency domain information of images, including both high and low frequencies, as the consistency constraint. Secondly, we incorporate supervision in the Encoder stage of model training to ensure that the model does not fail to learn due to the disruption of the original feature space caused by strong augmentation. Finally, extensive experimentation validates the effectiveness of our semi-supervised approach.

Machine Learning and Political Events: Application of a Semi-supervised Approach to Produce a Dataset on Presidential Cabinets

Machine Learning and Political Events: Application of a Semi-supervised Approach to Produce a Dataset on Presidential Cabinets

Semi-supervised learning from small annotated data and large unlabeled data for fine-grained Participants, Intervention, Comparison, and Outcomes entity recognition.

Extracting PICO elements-Participants, Intervention, Comparison, and Outcomes-from clinical trial literature is essential for clinical evidence retrieval, appraisal, and synthesis. Existing approaches do not distinguish the attributes of PICO entities. This study aims to develop a named entity recognition (NER) model to extract PICO entities with fine granularities. Using a corpus of 2511 abstracts with PICO mentions from 4 public datasets, we developed a semi-supervised method to facilitate the training of a NER model, FinePICO, by combining limited annotated data of PICO entities and abundant unlabeled data. For evaluation, we divided the entire dataset into 2 subsets: a smaller group with annotations and a larger group without annotations. We then established the theoretical lower and upper performance bounds based on the performance of supervised learning models trained solely on the small, annotated subset and on the entire set with complete annotations, respectively. Finally, we evaluated FinePICO on both the smaller annotated subset and the larger, initially unannotated subset. We measured the performance of FinePICO using precision, recall, and F1. Our method achieved precision/recall/F1 of 0.567/0.636/0.60, respectively, using a small set of annotated samples, outperforming the baseline model (F1: 0.437) by more than 16%. The model demonstrates generalizability to a different PICO framework and to another corpus, which consistently outperforms the benchmark in diverse experimental settings (P-value < .001). We developed FinePICO to recognize fine-grained PICO entities from text and validated its performance across diverse experimental settings, highlighting the feasibility of using semi-supervised learning (SSL) techniques to enhance PICO entities extraction. Future work can focus on optimizing SSL algorithms to improve efficiency and reduce computational costs. This study contributes a generalizable and effective semi-supervised approach leveraging large unlabeled data together with small, annotated data for fine-grained PICO extraction.

A semi-supervised boundary segmentation network for remote sensing images

Accurately segmenting remote sensing images remains challenging due to the diverse target scales and ambiguous structural boundaries. In this work, we propose a semi-supervised boundary segmentation network (BS-GAN) to address these challenges. BS-GAN employs a semi-supervised learning approach to reduce dependency on labeled data while introducing a novel mixed attention (MA) mechanism to enhance segmentation accuracy by aggregating long-range contextual information. Additionally, we develop a Boundary Gating Module (BGM) to refine boundary segmentation through a multi-task learning strategy focused on boundary feature enhancement. Experimental results on three benchmark datasets demonstrate that BS-GAN achieves superior accuracy and generalization capabilities compared to existing segmentation networks.

Three-dimensional semi-supervised lumbar vertebrae region of interest segmentation based on MAE pre-training

Background: The annotation of the regions of interest (ROI) of lumbar vertebrae by radiologists for bone density assessment is a tedious and time-intensive task. However, deep learning (DL) methods for image segmentation has the potential to substitute manual annotations which can significantly improve the efficiency of clinical diagnostics. Objective: The paper proposes a semi-supervised three-dimensional (3D) segmentation method for the ROI of lumbar vertebrae by integrating the tube masking masked autoencoder (MAE) pre-training. Methods: The paper proposes a method that modifies the masking strategy of the original MAE pre-training network. And the pre-training network is only trained by images without segmentation labels, when the training is finished, the weights will be saved for segmentation tasks. In downstream tasks, a semi-supervised approach utilizing pseudo-label generation is employed for training. This method leverages a small amount of labeled data to achieve the segmentation of ROI of the lumbar vertebrae. Results: The experimental results demonstrate that under the condition of limited annotated data, the proposed network improves the dice coefficient by 5–7% and reduces the hausdorff distance by 0.2∼0.6 mm compared to using the UNetr network alone for segmentation. When compared to the conventional MAE, the tube masking MAE presented in this paper assists effectively in segmentation, resulting in a 2% increase in the dice coefficient and a 0.24 mm reduction in the hausdorff distance. Conclusion: Automatic segmentation of the ROI of the lumbar vertebrae helps to shorten the time for doctors to annotate vertebrae during clinical bone density examinations. The paper employs the tube masking MAE pre-trained model to effectively extract contextual information of the 3D lumbar vertebrae, combining it with a semi-supervised network leveraging pseudo-label generation for fine-tuning, which leads to effective 3D segmentation of the lumbar vertebrae.

SSSA: low data sentiment analysis using boosting semi-supervised approach and deep feature learning network

SSSA: low data sentiment analysis using boosting semi-supervised approach and deep feature learning network

Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach.

Performing automatic and standardized 4D TEE segmentation and mitral valve analysis is challenging due to the limitations of echocardiography and the scarcity of manually annotated 4D images. This work proposes a semi-supervised training strategy using pseudo labelling for MV segmentation in 4D TEE; it employs a Teacher-Student framework to ensure reliable pseudo-label generation. 120 4D TEE recordings from 60 candidates for MV repair are used. The Teacher model, an ensemble of three convolutional neural networks, is trained on end-systole and end-diastole frames and is used to generate MV pseudo-segmentations on intermediate frames of the cardiac cycle. The pseudo-annotated frames augment the Student model's training set, improving segmentation accuracy and temporal consistency. The Student outperforms individual Teachers, achieving a Dice score of 0.82, an average surface distance of 0.37 mm, and a 95% Hausdorff distance of 1.72 mm for MV leaflets. The Student model demonstrates reliable frame-by-frame MV segmentation, accurately capturing leaflet morphology and dynamics throughout the cardiac cycle, with a significant reduction in inference time compared to the ensemble. This approach greatly reduces manual annotation workload and ensures reliable, repeatable, and time-efficient MV analysis. Our method holds strong potential to enhance the precision and efficiency of MV diagnostics and treatment planning in clinical settings.

Deep CNN-based semi-supervised learning approach for identifying and segmenting corrosion in hydraulic steel and water resources infrastructure

The United States faces significant challenges due to corrosion, with its impact on military and civilian infrastructure incurring over $20 billion in annual maintenance costs. The damage due to corrosion is profound, threatening structural safety, reducing esthetic value, and leading to costly repairs. To mitigate these effects, the Unified Facilities Criteria and Unified Facilities Guidance Specifications advise the use of protective coatings on metal surfaces. Early corrosion detection is crucial for maintaining structural integrity and minimizing maintenance costs. Recent breakthroughs in artificial intelligence and deep learning, including accurate corrosion classification, have significantly revolutionized the detection and management of corrosion. Despite these advancements, automatic corrosion segmentation in civil infrastructure remains challenging due to the scarcity of images and the labor-intensive annotation process. Moreover, existing segmentation methods are unable to manage the complexities that come with high-resolution corrosion images. This paper proposes a novel, semi-supervised, convolutional neural network-based image segmentation method for the automatic identification and segmentation of corrosion on coated steel surfaces, using both unlabeled and labeled corrosion images and leveraging the mean teacher model. The proposed novel method involves three steps: (1) utilizing high-resolution digital microscopy to capture detailed images and dividing them into manageable patches; (2) applying a semi-supervised learning approach, leveraging unlabeled corrosion images for enhanced segmentation precision; and (3) employing a smoothing module to improve the continuity of information. The proposed corrosion detection method has demonstrated promising performance with only 67% labeled data, achieving mean precision, recall, F-1 measure, and intersection over union of 90.0%, 96.2%, 92.7%, and 87.1%, respectively. Even with just 33% labeled data, the method maintains strong performance when compared to fully supervised deep learning models. This demonstrates a substantial data resource saving while ensuring accurate and reliable corrosion detection, which is crucial for infrastructure health monitoring. The successful validation of this approach provides a method that dramatically reduces the amount of visual data required to generate a reliable model.

Enhancing Specific Emitter Identification: A Semi-Supervised Approach with Deep Cloud and Broad Edge Integration

Enhancing Specific Emitter Identification: A Semi-Supervised Approach with Deep Cloud and Broad Edge Integration

Outlier detection in mixed-attribute data: a semi-supervised approach with fuzzy approximations and relative entropy

Outlier detection in mixed-attribute data: a semi-supervised approach with fuzzy approximations and relative entropy

Corrections to “A Semi-Supervised Learning Approach to Quality-Based Web Service Classification”

Corrections to “A Semi-Supervised Learning Approach to Quality-Based Web Service Classification”

A Novel Semi-Supervised Object Detection Approach via Scale Rebalancing and Global Proposal Contrast Consistency

A Novel Semi-Supervised Object Detection Approach via Scale Rebalancing and Global Proposal Contrast Consistency

Unmasking deception: a topic-oriented multimodal approach to uncover false information on social media

In the digital landscape, social media has emerged as a prevalent channel for global communication, connecting like-minded individuals worldwide. However, while facilitating information exchange, it is also susceptible to the dissemination of false information, posing a constant challenge to the reliability of online content. To address this issue, this paper introduces a novel methodology called TM-FID (Topic-oriented Multimodal False Information Detection), which combines false information detection and neural topic modeling within a semi-supervised multimodal approach. By jointly leveraging textual and visual information contained in online news, our approach provides insights into how false information influences specific discussion topics, thus enabling a comprehensive and fine-grained understanding of its spread and impact on social media conversation. Experimental evaluation carried out on a set of multimodal gossip-related news demonstrates the quality of the identified topics, assessed through a novel centroid-based metric, as well as the efficacy of the cross-attention mechanism used within TM-FID to accurately identify false information in multimodal news. Overall, the proposed methodology can enable effective strategies to counter the spread of false information, thereby fostering trust and confidence in the information shared on social media platforms.

Discrimination of high impedance fault in microgrid power network using semi-supervised machine learning algorithm

This work proposes a semi-supervised classification approach for discriminating high-impedance (HI) faults and other transients in a photovoltaic (PV) interconnected microgrid (MG) network. The suggested classifier combines unsupervised K-means clustering with the supervised multi-layer perceptron neural network algorithm. The K-means clustering technique is utilized in the first phase to detect and remove irrelevant instances from multiple events in the data set. To obtain the final predictions of targeted labels, clustered cases from the first phase are utilized to learn the multi-layer perceptron neural network classifier in the next phase. The suggested method outperforms stand-alone classifiers (K-means clustering and multi-layer perceptron) by providing enhanced accuracy and success rate of discriminating HI fault under standard test conditions and weather intermittency of PV. Furthermore, the results of the performance study clearly show that the suggested model is more resilient and offers superior performance than the stand-alone classifiers under the standard test condition and uncertainty of PV in MG networks.

Enhanced lithology identification with few-shot well-logging data using a Confidence-Enhanced Semi-Supervised Meta-Learning Approach

Enhanced lithology identification with few-shot well-logging data using a Confidence-Enhanced Semi-Supervised Meta-Learning Approach

Guidelines for cerebrovascular segmentation: Managing imperfect annotations in the context of semi-supervised learning.

Segmentation in medical imaging is an essential and often preliminary task in the image processing chain, driving numerous efforts towards the design of robust segmentation algorithms. Supervised learning methods achieve excellent performances when fed with a sufficient amount of labeled data. However, such labels are typically highly time-consuming, error-prone and expensive to produce. Alternatively, semi-supervised learning approaches leverage both labeled and unlabeled data, and are very useful when only a small fraction of the dataset is labeled. They are particularly useful for cerebrovascular segmentation, given that labeling a single volume requires several hours for an expert. In addition to the challenge posed by insufficient annotations, there are concerns regarding annotation consistency. The task of annotating the cerebrovascular tree is inherently ambiguous. Due to the discrete nature of images, the borders and extremities of vessels are often unclear. Consequently, annotations heavily rely on the expert subjectivity and on the underlying clinical objective. These discrepancies significantly increase the complexity of the segmentation task for the model and consequently impair the results. Consequently, it becomes imperative to provide clinicians with precise guidelines to improve the annotation process and construct more uniform datasets. In this article, we investigate the data dependency of deep learning methods within the context of imperfect data and semi-supervised learning, for cerebrovascular segmentation. Specifically, this study compares various state-of-the-art semi-supervised methods based on unsupervised regularization and evaluates their performance in diverse quantity and quality data scenarios. Based on these experiments, we provide guidelines for the annotation and training of cerebrovascular segmentation models.

Loss values of style transfer from UBM to AS-OCT images for plateau iris classification

Ultrasound biomicroscopy (UBM) is the standard for diagnosing plateau iris, but its limited accessibility in routine clinical settings presents challenges. While anterior segment optical coherence tomography (AS-OCT) is more convenient, its effectiveness in detecting plateau iris is limited. Previous research has demonstrated that combining UBM and AS-OCT image pairs through neural style transfer has improved classification accuracy. However, obtaining paired images is impractical in everyday practice. In this study, we propose a novel semi-supervised approach that eliminates the need for paired images. A generative model learns to distinguish plateau and non-plateau features from UBM images. AS-OCT images are input into the generator, which attempts to transform them into corresponding UBM images. The model’s performance is measured by loss values, representing the difficulty of transforming AS-OCT images, which are then used to predict plateau iris. The classification baseline, which applies AS-OCT solely without the style-transfer of UBM information, obtained 52.72% sensitivity, 60.82% specificity, and 57.89% accuracy for external validation; in contrast, the classification with neural style transfer of the image pairs respectively obtained 94.54%, 100.00%, and 98.03%, whereas the semi-supervised approach using loss values classification obtained 93.10%, 93.13%, and 93.12%, respectively. This semi-supervised transfer learning model presents a novel technique for detecting plateau iris with AS-OCT.

Deep-Transfer-Learning Strategies for Crop Yield Prediction Using Climate Records and Satellite Image Time-Series Data

The timely and reliable prediction of crop yields on a larger scale is crucial for ensuring a stable food supply and food security. In the last few years, many studies have demonstrated that deep learning can offer reliable solutions for crop yield prediction. However, a key challenge in applying deep-learning models to crop yield prediction is their reliance on extensive training data, which are often lacking in many parts of the world. To address this challenge, this study introduces TrAdaBoost.R2, along with fine-tuning and domain-adversarial neural network deep-transfer-learning strategies, for predicting the winter wheat yield across diverse climatic zones in the USA. All methods used the bidirectional LSTM (BiLSTM) architecture to leverage its sequential feature extraction capabilities. The proposed transfer-learning approaches outperformed the baseline deep-learning model, with mean absolute error reductions ranging from 9% to 28%, demonstrating the effectiveness of these methods. Furthermore, the results demonstrate that the semi-supervised transfer-learning approach using the two-stage version of TrAdaBoost.R2 and fine-tuning achieved a superior performance compared to the domain-adversarial neural network and standard TrAdaBoost.R2. Additionally, the study offers insights for improving the accuracy and generalizability of crop yield prediction models in diverse agricultural landscapes across different regions.

Semi-supervised Ensemble Learning for Automatic Interpretation of Lung Ultrasound Videos.

Point-of-care ultrasound (POCUS) stands as a safe, portable, and cost-effective imaging modality for swift bedside patient examinations. Specifically, lung ultrasonography (LUS) has proven useful in evaluating both acute and chronic pulmonary conditions. Despite its clinical value, automatic LUS interpretation remains relatively unexplored, particularly in multi-label contexts. This work proposes a novel deep learning (DL) framework tailored for interpreting lung POCUS videos, whose outputs are the finding(s) present in these videos (such as A-lines, B-lines, or consolidations). The pipeline, based on a residual (2+1)D architecture, initiates with a pre-processing routine for video masking and standardisation, and employs a semi-supervised approach to harness available unlabeled data. Additionally, we introduce an ensemble modeling strategy that aggregates outputs from models trained to predict distinct label sets, thereby leveraging the hierarchical nature of LUS findings. The proposed framework and its building blocks were evaluated through extensive experiments with both multi-class and multi-label models, highlighting its versatility. In a held-out test set, the categorical proposal, suited for expedite triage, achieved an average F1-score of 92.4%, while the multi-label proposal, helpful for patient management and referral, achieved an average F1-score of 70.5% across five relevant LUS findings. Overall, the semi-supervised methodology contributed significantly to improved performance, while the proposed hierarchy-aware ensemble provided moderate additional gains.