Self-supervised Representation Learning Research Articles

Improving inceptionV4 model based on fractional-order snow leopard optimization algorithm for diagnosing of ACL tears

In the current research study, a new method is presented to diagnose Anterior Cruciate Ligament (ACL) tears by introducing an optimized version of the InceptionV4 model. Our proposed methodology utilizes a custom-made variant of the Snow Leopard Optimization Algorithm, known as the Fractional-order Snow Leopard Optimization Algorithm (FO-LOA), to extract essential features from knee magnetic resonance imaging (MRI) images. This results in a substantial improvement in the accuracy of ACL tear detection. By effectively extracting critical features from knee MRI images, our proposed methodology significantly enhances diagnostic accuracy, potentially reducing false negatives and false positives. The enhanced model based on FO-LOA underwent thorough testing using the MRNet dataset, demonstrating exceptional performance metrics including an accuracy rate of 98.00%, sensitivity of 98.00%, precision of 97.00%, specificity of 98.00%, F1-score of 98.00%, and Matthews Correlation Coefficient (MCC) of 88.00%. These findings surpass current methodologies like Convolutional Neural Network (CNN), Inception-v3, Deep Belief Networks and Improved Honey Badger Algorithm (DBN/IHBA), integration of the CNN with an Amended Cooking Training-based Optimizer version (CNN/ACTO), Self-Supervised Representation Learning (SSRL), signifying a significant breakthrough in ACL injury diagnosis. Using FO-SLO to optimize the InceptionV4 framework shows promise in improving the accuracy of ACL tear identification, enabling prompt and efficient treatment interventions.

Advancements in Self-Supervised Learning for Remote Sensing Scene Classification: Present Innovations and Future Outlooks

Deep learning methodologies have significantly advanced the fields of computer vision and machine learning, enhancing performance across various tasks like classification, regression, and detection. In remote sensing for Earth observation, deep neural networks have propelled state-of-the-art results. However, a major drawback is their dependence on large annotated datasets, necessitating extensive human effort, especially in specialized domains like medical imaging or remote sensing. To mitigate this reliance on annotations, several self-supervised representation learning techniques have emerged, aiming to learn unsupervised image representations applicable to downstream tasks such as image classification, object detection, or semantic segmentation. Consequently, self-supervised learning approaches have gained traction in remote sensing. This article surveys the foundational principles of various self-supervised methods, focusing on scene classification tasks. We elucidate key contributions, analyze experimental setups, and synthesize findings from each study. Furthermore, we conduct comprehensive experiments on two public scene classification datasets to evaluate and benchmark different self-supervised models

Advancements in Self-Supervised Learning for Remote Sensing Scene Classification: Present Innovations and Future Outlooks

Deep learning methodologies have significantly advanced the fields of computer vision and machine learning, enhancing performance across various tasks like classification, regression, and detection. In remote sensing for Earth observation, deep neural networks have propelled state-of-the-art results. However, a major drawback is their dependence on large annotated datasets, necessitating extensive human effort, especially in specialized domains like medical imaging or remote sensing. To mitigate this reliance on annotations, several self-supervised representation learning techniques have emerged, aiming to learn unsupervised image representations applicable to downstream tasks such as image classification, object detection, or semantic segmentation. Consequently, self-supervised learning approaches have gained traction in remote sensing. This article surveys the foundational principles of various self-supervised methods, focusing on scene classification tasks. We elucidate key contributions, analyze experimental setups, and synthesize findings from each study. Furthermore, we conduct comprehensive experiments on two public scene classification datasets to evaluate and benchmark different self-supervised models

Self-Supervised Representation Learning on Electronic Health Records with Graph Kernel Infomax

Learning Electronic Health Records (EHRs) representation is a preeminent yet under-discovered research topic. It benefits various clinical decision support applications, e.g., medication outcome prediction or patient similarity search. Current approaches focus on task-specific label supervision on vectorized sequential EHR, which is not applicable to large-scale unsupervised scenarios. Recently, contrastive learning has shown great success in self-supervised representation learning problems. However, complex temporality often degrades the performance. We propose Graph Kernel Infomax, a self-supervised graph kernel learning approach on the graphical representation of EHR, to overcome the previous problems. Unlike the state-of-the-art, we do not change the graph structure to construct augmented views. Instead, we use Kernel Subspace Augmentation to embed nodes into two geometrically different manifold views. The entire framework is trained by contrasting nodes and graph representations on those two manifold views through the commonly used contrastive objectives. Empirically, using publicly available benchmark EHR datasets, our approach yields performance on clinical downstream tasks that exceeds the state-of-the-art. Theoretically, the variation in distance metrics naturally creates different views as data augmentation without changing graph structures. Practically, our method is non-ad hoc and confirms superior performance on commonly used graph benchmark datasets.

ProtoMGAE: Prototype-Aware Masked Graph Auto-Encoder for Graph Representation Learning

Graph self-supervised representation learning has gained considerable attention and demonstrated remarkable efficacy in extracting meaningful representations from graphs, particularly in the absence of labeled data. Two representative methods in this domain are graph auto-encoding and graph contrastive learning. However, the former methods primarily focus on global structures, potentially overlooking some fine-grained information during reconstruction. The latter methods emphasize node similarity across correlated views in the embedding space, potentially neglecting the inherent global graph information in the original input space. Moreover, handling incomplete graphs in real-world scenarios, where original features are unavailable for certain nodes, poses challenges for both types of methods. To alleviate these limitations, we integrate masked graph auto-encoding and prototype-aware graph contrastive learning into a unified model to learn node representations in graphs. In our method, we begin by masking a portion of node features and utilize a specific decoding strategy to reconstruct the masked information. This process facilitates the recovery of graphs from a global or macro level and enables handling incomplete graphs easily. Moreover, we treat the masked graph and the original one as a pair of contrasting views, enforcing the alignment and uniformity between their corresponding node representations at a local or micro level. Last, to capture cluster structures from a meso level and learn more discriminative representations, we introduce a prototype-aware clustering consistency loss that is jointly optimized with the preceding two complementary objectives. Extensive experiments conducted on several datasets demonstrate that the proposed method achieves significantly better or competitive performance on downstream tasks, especially for graph clustering, compared with the state-of-the-art methods, showcasing its superiority in enhancing graph representation learning.

SurgNet: Self-Supervised Pretraining With Semantic Consistency for Vessel and Instrument Segmentation in Surgical Images.

Blood vessel and surgical instrument segmentation is a fundamental technique for robot-assisted surgical navigation. Despite the significant progress in natural image segmentation, surgical image-based vessel and instrument segmentation are rarely studied. In this work, we propose a novel self-supervised pretraining method (SurgNet) that can effectively learn representative vessel and instrument features from unlabeled surgical images. As a result, it allows for precise and efficient segmentation of vessels and instruments with only a small amount of labeled data. Specifically, we first construct a region adjacency graph (RAG) based on local semantic consistency in unlabeled surgical images and use it as a self-supervision signal for pseudo-mask segmentation. We then use the pseudo-mask to perform guided masked image modeling (GMIM) to learn representations that integrate structural information of intraoperative objectives more effectively. Our pretrained model, paired with various segmentation methods, can be applied to perform vessel and instrument segmentation accurately using limited labeled data for fine-tuning. We build an Intraoperative Vessel and Instrument Segmentation (IVIS) dataset, comprised of ~3 million unlabeled images and over 4,000 labeled images with manual vessel and instrument annotations to evaluate the effectiveness of our self-supervised pretraining method. We also evaluated the generalizability of our method to similar tasks using two public datasets. The results demonstrate that our approach outperforms the current state-of-the-art (SOTA) self-supervised representation learning methods in various surgical image segmentation tasks.

Mining negative samples on contrastive learning via curricular weighting strategy

Contrastive learning, which pulls positive pairs closer and pushes away negative pairs, has remarkably propelled the development of self-supervised representation learning. Previous studies either neglected negative sample selection, resulting in suboptimal performance, or emphasized hard negative samples from the beginning of training, potentially leading to convergence issues. Drawing inspiration from curriculum learning, we find that learning with negative samples ranging from easy to hard improves both model performance and convergence rate. Therefore, we propose a dynamic negative sample weighting strategy for contrastive learning. Specifically, we design a loss function that adaptively adjusts the weights assigned to negative samples based on the model's performance. Initially, the loss prioritizes easy samples, but as training advances, it shifts focus to hard samples, enabling the model to learn more discriminative representations. Furthermore, to prevent an undue emphasis on false negative samples during later stages, which probably results in trivial solutions, we apply L2 regularization on the weights of hard negative samples. Extensive qualitative and quantitative experiments demonstrate the effectiveness of the proposed weighting strategy. The ablation study confirms both the reasonableness of the curriculum and the effectiveness of the regularization.

Federated Momentum Contrastive Clustering

Self-supervised representation learning and deep clustering are mutually beneficial to learn high-quality representations and cluster data simultaneously in centralized settings. However, it is not always feasible to gather large amounts of data at a central entity, considering data privacy requirements and computational resources. Federated Learning (FL) has been developed successfully to aggregate a global model while training on distributed local data, respecting the data privacy of edge devices. However, most FL research effort focuses on supervised learning algorithms. A fully unsupervised federated clustering scheme has not been considered in the existing literature. We present federated momentum contrastive clustering (FedMCC), a generic federated clustering framework that can not only cluster data automatically but also extract discriminative representations training from distributed local data over multiple users. In FedMCC, we demonstrate a two-stage federated learning paradigm where the first stage aims to learn differentiable instance embeddings and the second stage accounts for clustering data automatically. The experimental results show that FedMCC not only achieves superior clustering performance but also outperforms several existing federated self-supervised methods for linear evaluation and semi-supervised learning tasks. Additionally, FedMCC can easily be adapted to ordinary centralized clustering through what we call momentum contrastive clustering (MCC). We show that MCC achieves state-of-the-art clustering accuracy results in certain datasets such as STL-10 and ImageNet-10. We also present a method to reduce the memory footprint of our clustering schemes.

Generation-based Multi-view Contrast for Self-supervised Graph Representation Learning

Graph contrastive learning has made remarkable achievements in the self-supervised representation learning of graph-structured data. By employing perturbation function (i.e., perturbation on the nodes or edges of graph), most graph contrastive learning methods construct contrastive samples on the original graph. However, the perturbation-based data augmentation methods randomly change the inherent information (e.g., attributes or structures) of the graph. Therefore, after nodes embedding on the perturbed graph, we cannot guarantee the validity of the contrastive samples as well as the learned performance of graph contrastive learning. To this end, in this article, we propose a novel generation-based multi-view contrastive learning framework (GMVC) for self-supervised graph representation learning, which generates the contrastive samples based on our generator rather than perturbation function. Specifically, after nodes embedding on the original graph we first employ random walk in the neighborhood to develop multiple relevant node sequences for each anchor node. We then utilize the transformer to generate the representations of relevant contrastive samples of anchor node based on the features and structures of the sampled node sequences. Finally, by maximizing the consistency between the anchor view and the generated views, we force the model to effectively encode graph information into nodes embeddings. We perform extensive experiments of node classification and link prediction tasks on eight benchmark datasets, which verify the effectiveness of our generation-based multi-view graph contrastive learning method.

Non-parametric Representation Learning with Kernels

Unsupervised and self-supervised representation learning has become popular in recent years for learning useful features from unlabelled data. Representation learning has been mostly developed in the neural network literature, and other models for representation learning are surprisingly unexplored. In this work, we introduce and analyze several kernel-based representation learning approaches: Firstly, we define two kernel Self-Supervised Learning (SSL) models using contrastive loss functions and secondly, a Kernel Autoencoder (AE) model based on the idea of embedding and reconstructing data. We argue that the classical representer theorems for supervised kernel machines are not always applicable for (self-supervised) representation learning, and present new representer theorems, which show that the representations learned by our kernel models can be expressed in terms of kernel matrices. We further derive generalisation error bounds for representation learning with kernel SSL and AE, and empirically evaluate the performance of these methods in both small data regimes as well as in comparison with neural network based models.

Self-Supervised Representation Learning with Meta Comprehensive Regularization

Self-Supervised Learning (SSL) methods harness the concept of semantic invariance by utilizing data augmentation strategies to produce similar representations for different deformations of the same input. Essentially, the model captures the shared information among multiple augmented views of samples, while disregarding the non-shared information that may be beneficial for downstream tasks. To address this issue, we introduce a module called CompMod with Meta Comprehensive Regularization (MCR), embedded into existing self-supervised frameworks, to make the learned representations more comprehensive. Specifically, we update our proposed model through a bi-level optimization mechanism, enabling it to capture comprehensive features. Additionally, guided by the constrained extraction of features using maximum entropy coding, the self-supervised learning model learns more comprehensive features on top of learning consistent features. In addition, we provide theoretical support for our proposed method from information theory and causal counterfactual perspective. Experimental results show that our method achieves significant improvement in classification, object detection and semantic segmentation tasks on multiple benchmark datasets.

DDAE: Towards Deep Dynamic Vision BERT Pretraining

Recently, masked image modeling (MIM) has demonstrated promising prospects in self-supervised representation learning. However, existing MIM frameworks recover all masked patches equivalently, ignoring that the reconstruction difficulty of different patches can vary sharply due to their diverse distance from visible patches. In this paper, we propose a novel deep dynamic supervision to enable MIM methods to dynamically reconstruct patches with different degrees of difficulty at different pretraining phases and depths of the model. Our deep dynamic supervision helps to provide more locality inductive bias for ViTs especially in deep layers, which inherently makes up for the absence of local prior for self-attention mechanism. Built upon the deep dynamic supervision, we propose Deep Dynamic AutoEncoder (DDAE), a simple yet effective MIM framework that utilizes dynamic mechanisms for pixel regression and feature self-distillation simultaneously. Extensive experiments across a variety of vision tasks including ImageNet classification, semantic segmentation on ADE20K and object detection on COCO demonstrate the effectiveness of our approach.

SwitchTab: Switched Autoencoders Are Effective Tabular Learners

Self-supervised representation learning methods have achieved significant success in computer vision and natural language processing (NLP), where data samples exhibit explicit spatial or semantic dependencies. However, applying these methods to tabular data is challenging due to the less pronounced dependencies among data samples. In this paper, we address this limitation by introducing SwitchTab, a novel self-supervised method specifically designed to capture latent dependencies in tabular data. SwitchTab leverages an asymmetric encoder-decoder framework to decouple mutual and salient features among data pairs, resulting in more representative embeddings. These embeddings, in turn, contribute to better decision boundaries and lead to improved results in downstream tasks. To validate the effectiveness of SwitchTab, we conduct extensive experiments across various domains involving tabular data. The results showcase superior performance in end-to-end prediction tasks with fine-tuning. Moreover, we demonstrate that pre-trained salient embeddings can be utilized as plug-and-play features to enhance the performance of various traditional classification methods (e.g., Logistic Regression, XGBoost, etc.). Lastly, we highlight the capability of SwitchTab to create explainable representations through visualization of decoupled mutual and salient features in the latent space.

VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding

Vision and language foundation models (VLMs) have showcased impressive capabilities in 2D scene understanding. However, their latent potential in elevating the understanding of 3D autonomous driving scenes remains untapped. In this paper, we propose VLM2Scene, which exploits the potential of VLMs to enhance 3D self-supervised representation learning through our proposed image-text-LiDAR contrastive learning strategy. Specifically, in the realm of autonomous driving scenes, the inherent sparsity of LiDAR point clouds poses a notable challenge for point-level contrastive learning methods. This method often grapples with limitations tied to a restricted receptive field and the presence of noisy points. To tackle this challenge, our approach emphasizes region-level learning, leveraging regional masks without semantics derived from the vision foundation model. This approach capitalizes on valuable contextual information to enhance the learning of point cloud representations. First, we introduce Region Caption Prompts to generate fine-grained language descriptions for the corresponding regions, utilizing the language foundation model. These region prompts then facilitate the establishment of positive and negative text-point pairs within the contrastive loss framework. Second, we propose a Region Semantic Concordance Regularization, which involves a semantic-filtered region learning and a region semantic assignment strategy. The former aims to filter the false negative samples based on the semantic distance, and the latter mitigates potential inaccuracies in pixel semantics, thereby enhancing overall semantic consistency. Extensive experiments on representative autonomous driving datasets demonstrate that our self-supervised method significantly outperforms other counterparts. Codes are available at https://github.com/gbliao/VLM2Scene.

Abstract 4298: Predicting tumor evolution from digital histology using AI

Abstract Diverse clinical presentations of clear cell renal cell carcinoma (ccRCC) confound clinical decision making, leading to over and undertreatment. Clonal evolution of ccRCC proceeds through distinct trajectories characterised by differing levels of genomic intratumoral heterogeneity (gITH) and chromosomal complexity (weighted genomic instability index, wGII). However, accurate evaluation of these indices requires multiregional profiling of fresh tumour; cost prohibitive and logistically challenging in the clinical setting. Clinical histopathology workflows routinely capture multiple tumour areas enabling the use artificial intelligence (AI) to predict tumour evolutionary features directly from clinical grade H&E whole slide image (WSIs). ccRCC displays profound genetic and histological ITH but the link between these entities remains unclear. We leverage the TRACERx Renal cohort, comprising 1485 WSIs from 81 tumours to predict WGII and gITH and to gain insights into the relationship between genetic and histological ITH. Critically, each WSI is associated with a wGII and gITH label derived from a closely linked fresh tumour sample. For both prediction tasks, we extracted meaningful features for each WSI using self-supervised representation learning “MoCo”. Since high wGII confers poor prognosis we focussed on predicting binary stratification label of high wGII or low wGII (relative to the cohort median). First we predicted wGII as a continuous variable using a supervised multiple instance learning regression model trained on the MoCo features, and then classified the predicted wGII into “high” or “low” achieving 0.80 AUROC. To predict gITH we postulated that the degree of gITH would correlate with histological ITH. Using an unsupervised clustering of refined MoCo features we defined 24 histological clusters. The number of computationally derived histological clusters within a single tumour positively correlated with gITH (pearson’s 0.56). We used the number of clusters to classify WSIs into prognostic binary groups of high or low gITH (relative to the cohort median) achieving an AUROC of 0.80. To understand the biological relationship between histological and genetic ITH we pathologically characterised the histological clusters: a pathologist annotated WSIs with tumour architecture and cytomorphology. Image tiles were associated with the annotations using spatial coordinates, illuminating phenotypic traits of different evolutionary trajectories and providing an interpretability framework for our AI pipelines. Since the tumour evolutionary course dictates disease progression tempo, applying evolutionary classification in clinic can fundamentally improve patient care. Here, for the first time, we provide a framework to translate fundamental evolutionary principles underpinning tumour biology and clinical progression into a prognostic computational pathology biomarker possible to clinically implement. Citation Format: Charlotte E. Spencer, Axel Camara, Auriane Riou, Lewis Au, Jose I. Lopez, Zayd Tippu, Charles Maussion, Kenneth Ho, Amy Strange, Emma Nye, Veronique Birault, Lydwine Van-praet, Kim Edmonds, Eleanor Carlyle, Steve Hazell, Sarah Rudman, James Larkin, Samra Turajlic. Predicting tumor evolution from digital histology using AI [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4298.

3D seismic Fault Detection via Contrastive-Reconstruction Representation Learning

Fault detection is a critical step in structural modeling and characterizing reservoirs. 3D seismic fault labeling is almost impossible to obtain, while networks trained on synthetic fault data have limited generalization to real data. We use self-supervised representation learning to enable networks to learn seismic field data features during pre-training, improving generalization. However, widely popular ViT/SwinViT-based methods cannot capture low-level fault features well. CNN-based have limited capability in transferring to downstream tasks. Tackle this, we designed the Tiny Self-Attention and extensively embedded it into the HRNet. It merges the advantages of convolutional and self-attention, allowing for in-depth learning of seismic representation information during the representation learning phase, thus achieving better inter-class separation in downstream tasks. We crafted two proxy tasks. The contrastive task aims to minimize the projected feature distance of overlapping areas from two distinct views. To tackle the issue of memory overflow and training suspension caused by the high spatial and temporal complexity of 3D high-resolution feature matching, we introduced sparse distance matching and an adaptive feature aggregation. In reconstruction task, we designed a masking strategy tailored to the unique attributes of 3D seismic data. It process named “Fault Detection via Contrast-Reconstruction Representation Learning” (FaultCRL). Experimentally, we compared the latest fault detection methods, and representation learning techniques like MAE and SwinUNETR, showcasing the notable advantage of FaultCRL in fault detection tasks. The appendix provides all qualitative results from mainstream geophysical journals regarding The Netherlands-F3 data in recent years, indicating that our approach has reached the current state-of-the-art level.

Self-Supervised Representation Learning From Temporal Ordering of Automated Driving Sequences

Self-Supervised Representation Learning From Temporal Ordering of Automated Driving Sequences

Boundary-aware information maximization for self-supervised medical image segmentation

Self-supervised representation learning can boost the performance of a pre-trained network on downstream tasks for which labeled data is limited. A popular method based on this paradigm, known as contrastive learning, works by constructing sets of positive and negative pairs from the data, and then pulling closer the representations of positive pairs while pushing apart those of negative pairs. Although contrastive learning has been shown to improve performance in various classification tasks, its application to image segmentation has been more limited. This stems in part from the difficulty of defining positive and negative pairs for dense feature maps without having access to pixel-wise annotations.In this work, we propose a novel self-supervised pre-training method that overcomes the challenges of contrastive learning in image segmentation. Our method leverages Information Invariant Clustering (IIC) as an unsupervised task to learn a local representation of images in the decoder of a segmentation network, but addresses three important drawbacks of this approach: (i) the difficulty of optimizing the loss based on mutual information maximization; (ii) the lack of clustering consistency for different random transformations of the same image; (iii) the poor correspondence of clusters obtained by IIC with region boundaries in the image. Toward this goal, we first introduce a regularized mutual information maximization objective that encourages the learned clusters to be balanced and consistent across different image transformations. We also propose a boundary-aware loss based on cross-correlation, which helps the learned clusters to be more representative of important regions in the image. Compared to contrastive learning applied in dense features, our method does not require computing positive and negative pairs and also enhances interpretability through the visualization of learned clusters.Comprehensive experiments involving four different medical image segmentation tasks reveal the high effectiveness of our self-supervised representation learning method. Our results show the proposed method to outperform by a large margin several state-of-the-art self-supervised and semi-supervised approaches for segmentation, reaching a performance close to full supervision with only a few labeled examples.

Self-Adaptive Training: Bridging Supervised and Self-Supervised Learning.

We propose self-adaptive training-a unified training algorithm that dynamically calibrates and enhances training processes by model predictions without incurring an extra computational cost-to advance both supervised and self-supervised learning of deep neural networks. We analyze the training dynamics of deep networks on training data that are corrupted by, e.g., random noise and adversarial examples. Our analysis shows that model predictions are able to magnify useful underlying information in data and this phenomenon occurs broadly even in the absence of any label information, highlighting that model predictions could substantially benefit the training processes: self-adaptive training improves the generalization of deep networks under noise and enhances the self-supervised representation learning. The analysis also sheds light on understanding deep learning, e.g., a potential explanation of the recently-discovered double-descent phenomenon in empirical risk minimization and the collapsing issue of the state-of-the-art self-supervised learning algorithms. Experiments on the CIFAR, STL, and ImageNet datasets verify the effectiveness of our approach in three applications: classification with label noise, selective classification, and linear evaluation. To facilitate future research, the code has been made publicly available at https://github.com/LayneH/self-adaptive-training.

Self-supervised representation learning of filtration barrier in kidney

While the advance of deep learning has allowed to automate many tasks in bioimage analysis, quantifying key visual features of biological objects in an image, such as cells, organs, or tissues, is still a multi-step and laborious task. It requires image segmentation and definition of features of interest, which often might be image- and problem-specific. This approach requires image labeling and training of the segmentation method as well as manual feature design and implementation of dedicated procedures for their quantification. Here we propose a self-supervised learning (SSL) approach to encoding in microscopy images morphological features of molecular structures that play role in disease phenotype and patient clinical diagnosis. We encode super-resolution images of slit diaphragm (SD)—a specialized membrane between podocyte cells in kidney—in a high-dimensional embedding space in an unsupervised manner, without the need of image segmentation and feature quantification. We inspect the embedding space and demonstrate its relationship to the morphometric parameters of the SD estimated with a previously published method. The SSL-derived image representations additionally reflect the level of albuminuria—a key marker of advancement of kidney disease in a cohort of chronic kidney disease patients. Finally, the embeddings allow for distinguishing mouse model of kidney disease from the healthy subjects with a comparable accuracy to classification based on SD morphometric features. In a one step and label-free manner the SSL approach offers possibility to encode meaningful details in biomedical images and allow for their exploratory, unsupervised analysis as well as further fine-tuning for specialized supervised tasks.