Self-supervised Learning via Inter-modal Reconstruction and Feature Projection Networks for Label-Efficient 3D-to-2D Segmentation

Spatiotemporal self-supervised predictive learning for atmospheric variable prediction via multi-group multi-attention

Considering several patients screened due to COVID-19 pandemic, computer-aided detection has strong potential in assisting clinical workflow efficiency and reducing the incidence of infections among radiologists and healthcare providers. Since many confirmed COVID-19 cases present radiological findings of pneumonia, radiologic examinations can be useful for fast detection. Therefore, chest radiography can be used to fast screen COVID-19 during the patient triage, thereby determining the priority of patient's care to help saturated medical facilities in a pandemic situation. In this paper, we propose a new learning scheme called self-supervised transfer learning for detecting COVID-19 from chest X-ray (CXR) images. We compared six self-supervised learning (SSL) methods (Cross, BYOL, SimSiam, SimCLR, PIRL-jigsaw, and PIRL-rotation) with the proposed method. Additionally, we compared six pretrained DCNNs (ResNet18, ResNet50, ResNet101, CheXNet, DenseNet201, and InceptionV3) with the proposed method. We provide quantitative evaluation on the largest open COVID-19 CXR dataset and qualitative results for visual inspection. Our method achieved a harmonic mean (HM) score of 0.985, AUC of 0.999, and four-class accuracy of 0.953. We also used the visualization technique Grad-CAM++ to generate visual explanations of different classes of CXR images with the proposed method to increase the interpretability. Our method shows that the knowledge learned from natural images using transfer learning is beneficial for SSL of the CXR images and boosts the performance of representation learning for COVID-19 detection. Our method promises to reduce the incidence of infections among radiologists and healthcare providers.

Highlights Convolutional neural network and MMD transfer learning methods are applied in inversion of optical properties. The classification accuracy of apples’ peel and pulp absorption coefficients are 84.61% and 92.47%, the accuracy of peel and pulp scattering coefficients are 83.56% and 86.53%, respectively. The depth optical characteristics can better reflect brix and moisture of apple then optical properties and hyperspectral data, the correlations are in the form of 0.98 and 0.98. Abstract. An inversion of optical properties is an important test for determining the quality of fruit. The conventional inversion model of the optical properties uses measured hyperspectral images as the training data. Studies show that the conventional machine learning method for inverting the optical properties results in low inversion accuracy, especially with curved models. Hence, the present study uses a convolutional neural network scheme to train the simulated hyperspectral images. Moreover, the maximum mean discrepancy (MMD) transfer method is used to transfer the simulated hyperspectral images to the measured hyperspectral images of apples. To evaluate the performance of the proposed method, the present study uses it to classify a variety of an apple’s optical properties, including the peel absorption, pulp absorption, peel scattering, and pulp scattering coefficients. The classification accuracies of the peel and pulp absorption coefficients are 84.61% and 92.47%, respectively. The classification accuracies of the peel and pulp scattering coefficients are 83.56% and 86.53%, respectively. These inversion results are compared with convolutional neural networks, neural networks, and support vector machines with measured hyperspectral images. It was found that the proposed inversion model is an effective scheme for optical property inversion. To prove the necessity of optical property inversion, the least squares, decision tree and random forest regression methods are performed to analyze the correlation between the depth of optical characteristics and the brix and moisture. The present study shows that these correlations are in the form of 0.98 and 0.98. The correlation coefficients increase by 0.36 and 0.25 compared to the measured hyperspectral images. The conclusions show that the proposed inversion model is an effective scheme for apple optical property inversion. Keywords: Apple tissue, Hyperspectral, Optical property inversion, Quality inspection.

This review provides a comprehensive overview of the advancements in deep learning for electroencephalography (EEG) analysis. EEG offers a non-invasive means to monitor brain activity, with applications in brain-computer interfaces (BCIs), clinical diagnostics, emotion recognition, and sleep monitoring. We trace the historical progression from early convolutional neural networks (CNNs) in the mid-2010s to the recent emergence of transformer-based models. Key CNN architectures, such as EEGNet, are examined for their efficiency in feature extraction from raw EEG data. Transformer and hybrid models are discussed, highlighting their ability to model long-range dependencies and achieve state-of-the-art performance across diverse tasks like motor imagery, seizure detection, and sleep staging. The review also addresses self-supervised and unsupervised learning methods to mitigate data scarcity. Challenges including inter-subject variability, noise artifacts, and model interpretability are analyzed, alongside future directions such as transfer learning, synthetic data generation, domain adaptation, and multi-modal fusion. By synthesizing these developments, this work aims to guide future research toward more generalizable and practical EEG deep learning systems.

Self-supervised learning (SSL) methods aim to learn view-invariant representations by maximizing the similar-ity between the features extracted from different crops of the same image regardless of cropping size and content. In essence, this strategy ignores the fact that two crops may truly contain different image information, e.g., background and small objects, and thus tends to restrain the diversity of the learned representations. In this work, we address this issue by introducing a new self-supervised learning strat-egy, LoGo, that explicitly reasons about Local and Global crops. To achieve view invariance, LoGo encourages similarity between global crops from the same image, as well as between a global and a local crop. However, to correctly encode the fact that the content of smaller crops may differ entirely, LoGo promotes two local crops to have dissimi-lar representations, while being close to global crops. Our LoGo strategy can easily be applied to existing SSL meth-ods. Our extensive experiments on a variety of datasets and using different self-supervised learning frameworks vali-date its superiority over existing approaches. Noticeably, we achieve better results than supervised models on trans-fer learning when using only 1/10 of the data. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> Our code and pretrained models can be found at https://github.com/ztt1024/LoGo-SSL.

Adaptive self-supervised learning for sequential recommendation

The implementation of computer vision generally focuses on resolving a problem of one of two types of image data, whether it was natural image datasets or medical image datasets. Self-supervised learning (SSL), as the most recent approach in computer vision, is typically used to solved tasks related only to natural image datasets such as ImageNet, CIFAR-10, and PASCAL. This study focuses on analyzing the involvement of natural images in the SSL implementation when classifying medical images. We used the ResNet50 architecture comparing the results of human embryo image classification using the conventional transfer learning method and SSL methods. We apply three SSL approaches, namely Rotation, Jigsaw Puzzle, and SimCLRv1. The results of our research, with 1226 embryo images divided into three classes, show that the SSL implementations cannot be said to be efficient enough compared to the conventional model produced by ResNet50 with pretrained weight ImageNet. However, when the models were trained using random weight initialization, the Rotation model with an average accuracy of 78.15% and the Jigsaw Puzzle model with an average accuracy of 80.30% are significantly superior to the conventional model with average accuracy is 71.07%. Moreover, related to our embryos dataset, we discuss the differences between the SSL context-context contrastive learning approach and the context instance contrastive learning approach.

Deep models trained in supervised mode have achieved remarkable success on a variety of tasks. When labeled samples are limited, self-supervised learning (SSL) is emerging as a new paradigm for making use of large amounts of unlabeled samples. SSL has achieved promising performance on natural language and image learning tasks. Recently, there is a trend to extend such success to graph data using graph neural networks (GNNs). In this survey, we provide a unified review of different ways of training GNNs using SSL. Specifically, we categorize SSL methods into contrastive and predictive models. In either category, we provide a unified framework for methods as well as how these methods differ in each component under the framework. Our unified treatment of SSL methods for GNNs sheds light on the similarities and differences of various methods, setting the stage for developing new methods and algorithms. We also summarize different SSL settings and the corresponding datasets used in each setting. To facilitate methodological development and empirical comparison, we develop a standardized testbed for SSL in GNNs, including implementations of common baseline methods, datasets, and evaluation metrics.

The field of machine learning has recently undergone significant advancements, driven by the development of sophisticated algorithms and the availability of large datasets. Among the various paradigms of machine learning, supervised learning has traditionally dominated, relying heavily on labeled data to train models. However, acquiring and annotating large amounts of labeled data is both time-consuming and costly, leading to a growing interest in self-supervised learning (SSL). This promising approach leverages unlabeled data to learn useful representations, which can then be fine-tuned for specific tasks with minimal labeled data Self-supervised learning creates learning signals from the data itself, thus eliminating the need for manual labeling. By solving auxiliary tasks, such as predicting the rotation angle of an image or filling in missing parts of data, SSL methods can learn robust feature representations that are useful for a wide range of downstream tasks. This paradigm reduces dependency on labeled data and exploits the abundance of unlabeled data available in various domains. The significance of SSL lies in its potential to democratize machine learning, making it more accessible and applicable across different fields. From natural language processing to computer vision and beyond, self-supervised techniques have demonstrated remarkable success in improving model performance while reducing reliance on extensive labeled datasets. As such, SSL represents a critical step towards more efficient and scalable machine learning solutions. This paper explores the development and implementation of a novel self-supervised learning solution based on machine learning principles. The proposed method achieves an accuracy of 97.6%, a mean absolute error (MAE) of 0.403, and a root mean square error (RMSE) of 0.203. A comprehensive overview of the underlying methodologies, the challenges faced, and the potential applications of this approach is provided. By examining the theoretical foundations and practical implementations, this study aims to contribute to the growing body of research in SSL and highlight its transformative impact on the machine learning landscape.Self-Supervised Learning.

The artificial intelligence (AI) community has recently made tremendous progress in developing self-supervised learning (SSL) algorithms that can learn high-quality data representations from massive amounts of unlabeled data. These methods brought great results even to the fields outside of AI. Due to the joint efforts of researchers in various areas, new SSL methods come out daily. However, such a sheer number of publications make it difficult for beginners to see clearly how the subject progresses. This survey bridges this gap by carefully selecting a small portion of papers that we believe are milestones or essential work. We see these researches as the “dots” of SSL and connect them through how they evolve. Hopefully, by viewing the connections of these dots, readers will have a high-level picture of the development of SSL across multiple disciplines including natural language processing, computer vision, graph learning, audio processing, and protein learning.

Recent research progress in shadow detection has leveraged the development of remote sensing and computer vision. Since shadows of buildings, trees, bridges in one image can provide useful information about the scene to help people understand the shape, feature or estimate their locations and orientations of original objects, especially for damaged objects. In this study, a novel shadow detection algorithm for remote sensing imagery, called self-supervised learning method is proposed. The aim of this work is to generate shadow ratio threshold automatically without human interaction. To alleviate the traditional drawbacks of shadow detection, we fully combine supervised and unsupervised shadow detection method to suggest a self-supervised learning method, which supports us a strongly clue with establishing the relation of shadow and its original object. Subsequently, we benefit from gray-scale histogram to extract shadow segments, then shadow outlines are obtained. Finally, we assess the shadow detection performance of the proposed approach by comparing our results with the state-of-the-art methods. The results reveal the applicability and precision of the proposed self-supervised learning shadow detection technique.

Self-supervised learning provides a possible solution to extract effective visual representations from unlabeled histopathological images. However, existing methods either fail to make good use of domain-specific knowledge, or rely on side information like spatial proximity and magnification. In this paper, we propose CS-CO, a hybrid self-supervised visual representation learning method tailored for histopathological images, which integrates advantages of both generative and discriminative models. The proposed method consists of two self-supervised learning stages: cross-stain prediction (CS) and contrastive learning (CO), both of which are designed based on domain-specific knowledge and do not require side information. A novel data augmentation approach, stain vector perturbation, is specifically proposed to serve contrastive learning. Experimental results on the public dataset NCT-CRC-HE-100K demonstrate the superiority of the proposed method for histopathological image visual representation. Under the common linear evaluation protocol, our method achieves 0.915 eight-class classification accuracy with only 1,000 labeled data, which is about 1.3% higher than the fully-supervised ResNet18 classifier trained with the whole 89,434 labeled training data. Our code is available at https://github.com/easonyang1996/CS-CO.

A limited amount of training data is a well-known challenge in the application of deep learning methods. This issue is particularly relevant in biomedical signal processing, such as the analysis of electrocardiograms, due to the labor-intensive nature of data preparation, which requires the involvement of qualified specialists. Self-supervised learning methods, originally developed in such domains as natural language processing and computer vision, have emerged as a potential approach to addressing this challenge and are increasingly being explored in biomedical signal processing. However, direct adaptation of self-supervised learning techniques from other domains does not fully account for ECG-specific characteristics, such as quasi-periodicity, localized morphological features, and susceptibility to noise. This highlights the relevance of developing ECG-specific self-supervised learning methods. This study presents a novel self-supervised learning approach for pretraining neural networks on unlabeled ECG data. The proposed method is based on predicting the short consecutive signal segment using a preceding one and a learned representation vector. The representation extraction and prediction models are trained jointly on the MIMIC-ECG-IV dataset using backpropagation to minimize the mean squared error between the predicted and original signal segments. As an example of a downstream task, a linear binary classifier was trained on the PTB-XL dataset to diagnose pathological conditions using Lead I. The number of training examples for each diagnosis was limited to thirty-four samples. Firstly, the representation model was pre-trained on the unlabeled MIMIC-ECG-IV dataset, and then linear classifiers were trained on the learned representations for each selected diagnosis in PTB-XL. A comparison was also conducted with a randomly initialized representation model trained jointly with the classifier in a fully supervised manner. The proposed method was evaluated against adaptations of Contrastive Learning, Contrastive Predictive Coding, and Masked Autoencoders method. To ensure a controlled experimental setup, implementations of all considered methods were developed using a unified codebase and shared architectural components. Experimental results demonstrated a significant advantage of all self-supervised learning approaches over joint training of feature extraction and classification models. The proposed SSL method outperformed other tested approaches, particularly for diagnoses with subtle short-term morphological features, such as atrial fibrillation and flutter. These findings suggest the potential for further research in developing ECG-specific self-supervised learning methods as a promising approach to improving neural network performance in scenarios with limited labeled data.

We read with great interest the study by Kumar et al. [1] on the detection of reticular pseudodrusen (RPD) using a deep learning (DL) model on optical coherence tomography (OCT). The authors' comprehensive validation on multiple independent external datasets is commendable. Their work represents a significant step toward the automatic detection of this key high-risk phenotype in age-related macular degeneration (AMD). However, we wish to highlight three points to further evaluate the robustness and clinical applicability of the model. First, the reliance on a “single observer” for the entire training dataset (9800 B-scans) raises concerns regarding potential systematic bias. In medical image analysis, interobserver variability is substantial for subtle radiographic features such as RPD (intergrader Dice similarity coefficient [DSC] = 0.68 in this study). Consequently, relying on a single expert's criteria could systematically introduce subjective judgements into the model training, potentially codifying grader-specific biases as demonstrated in other retinal deep learning applications [2]. The finding that the model-grader agreement (DSC = 0.76) exceeded the intergrader agreement is noteworthy but warrants caution: does this reflect superior objective accuracy, or does it suggest the model has successfully “overfitted” to the personal interpretation style of that specific expert? Future iterations should prioritise consensus grading from multiple experts to mitigate annotation noise and ensure the model learns generalised pathological features. Second, while the authors suggest the model holds potential for quantifying RPD extent to monitor disease progression, the external validation was limited to qualitative detection (presence/absence). Given that the clinical significance of RPD is closely tied to its area expansion and association with geographic atrophy growth [3], validating the segmentation performance in external cohorts is essential. We suggest the authors, if data permit, perform a supplementary analysis reporting segmentation metrics (e.g., Dice scores) on the external validation sets. This would provide stronger evidence for the model's utility in longitudinal monitoring. Finally, the model was developed and validated exclusively using Heidelberg Spectralis images. In real-world clinical settings, variability across devices (e.g., Zeiss or Topcon) can lead to performance degradation due to domain shift [4], a well-established vulnerability of medical AI models trained on single-source data. Therefore, the model's generalisability to other common imaging platforms remains to be established. Future research utilising multi-vendor data would significantly enhance its applicability across diverse clinical scenarios. In summary, this study provides a foundational tool for automated RPD detection. Addressing these points regarding ground truth consensus, quantitative validation, and cross-vendor universality will further solidify its role in the clinical management of AMD. The authors have nothing to report. The authors declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

The objective of this study is to develop a computer-aided diagnosis (CADx) system for successful ultrasound-guided supraclavicular block (SCB). The retrospectively collected ultrasound videos were from 800 patients to develop the CADx system (600 to the training and validation set, and two test sets of 100 each). The proposed method consists of classification and segmentation approaches using convolutional neural networks (CNN). As for the classification method, a ResNet-based model using augmentation technique, GRU module, and self-supervised learning method were added to the comparison experiment. The segmentation approach used ResNet as encoder for U-Net, and it is a cascaded structure that trains the classification model successively by using the prediction result of the U-Net as pseudo labels. As a result of the classification and segmentation approaches, the ResNet layer did not improve performance further in layers deeper than 34, but applying the augmentation methods was effective. In addition, it was confirmed that the classification approach improved performance when the GRU modules were added, but was not suitable for real-time setting. The proposed approaches showed the highest performance with accuracy 0.88, 0.883, precision 0.578, 0.621, recall 0.712, 0.601, F1-score 0.639, 0.609, and AUROC 0.913, 0.919, respectively.