Modality-invariant Representations Research Articles

Discovering attention-guided cross-modality correlation for visible–infrared person re-identification

Visible–infrared person re-identification (VI Re-ID) is an essential and challenging task. Existing studies mainly focus on learning the unified modality-invariant representations directly from visible and infrared images. However, it is hard to obtain the identity-aware patterns due to the co-existence of inter- and intra-modality discrepancies. In this paper, we propose a novel attention-guided cross-modality correlation method (AGCC) to achieve the modality-invariant and identity-discriminative representations for visible–infrared person Re-ID. Specifically, we introduce a modality-aware attention (MAA) mechanism to model the inter- and intra-modality variations, which generates attention masks of two modalities for preserving the most significant region and obtaining the discriminative patterns in each identity. Further, we present an attention-guided channel and spatial correlation scheme (AGCSC) to establish the attention-guided cross-modality correlation, which can bridge the gap between inter- and intra-modalities. Moreover, a novel joint-modality learning head (JMLH) is developed to promote the metric and mutual learning from both feature distribution and classification logit levels. Extensive experiments on two public SYSU-MM01 and RegDB datasets demonstrate the remarkable superiority of our method over the state of the arts. The implementation codes will be made available soon.

MIT-FRNet: Modality-invariant temporal representation learning-based feature reconstruction network for missing modalities

The investigation of missing modalities aims to extract valuable feature information from missing multi-modal data, it is a focal point in multi-modal learning. Existing missing modalities processing methods primarily focused on multi-modal fusion schemes to achieve optimal performance, but they faced the following two key challenges: (1) how to improve the robustness of incomplete multi-modal sequence representation, (2) how to effectively learn modality-invariant representations to mitigate heterogeneity between modalities. In this paper, we propose a modality-invariant temporal representation learning-based feature reconstruction network called MIT-FRNet for the missing modalities to tackle these challenges. In the MIT-FRNet, we first extract the latent features for each modality considering intra-modality and inter-modality, and then introduce an encoder-decoder framework to address the first challenge, it reconstructs missing element features via taking the incomplete modal sequences as input and implementing the inter-modal and cross-modal attention mechanisms for feature extraction. And then, through treating each timestamp as a single Gaussian distribution, we design a fine-grained similarity constraint based on distribution-level modality-invariant representations to learn effective modality-invariant representations, thereby addressing the second challenge. Finally, the efficiency of proposed model is validated through the classification results after multi-modal fusion that involves using a gate encoder to pass it and followed by a vector fusion to fuse it. Extensive experiments on public benchmark datasets demonstrate that the proposed MIT-FRNet method achieves promising results under varying missing rates while exhibiting good convergence.

Incomplete Cross-Modal Retrieval with Deep Correlation Transfer

Most cross-modal retrieval methods assume the multi-modal training data is complete and has a one-to-one correspondence. However, in the real world, multi-modal data generally suffers from missing modality information due to the uncertainty of data collection and storage processes, which limits the practical application of existing cross-modal retrieval methods. Although some solutions have been proposed to generate the missing modality data using a single pseudo sample, this may lead to incomplete semantic restoration and sub-optimal retrieval results due to the limited semantic information it provides. To address this challenge, this article proposes an Incomplete Cross-Modal Retrieval with Deep Correlation Transfer (ICMR-DCT) method that can robustly model incomplete multi-modal data and dynamically capture the adjacency semantic correlation for cross-modal retrieval. Specifically, we construct intra-modal graph attention-based auto-encoder to learn modality-invariant representations by performing semantic reconstruction through intra-modality adjacency correlation mining. Then, we design dual cross-modal alignment constraints to project multi-modal representations into a common semantic space, thus bridging the heterogeneous modality gap and enhancing the discriminability of the common representation. We further introduce semantic preservation to enhance adjacency semantic information and achieve cross-modal semantic correlation. Moreover, we propose a nearest-neighbor weighting integration strategy with cross-modal correlation transfer to generate the missing modality data according to inter-modality mapping relations and adjacency correlations between each sample and its neighbors, which improves the robustness of our method against incomplete multi-modal training data. Extensive experiments on three widely tested benchmark datasets demonstrate the superior performance of our method in cross-modal retrieval tasks under both complete and incomplete retrieval scenarios. Our used datasets and source codes are available at https://github.com/shidan0122/DCT.git .

Deep Hierarchical Multimodal Metric Learning.

Multimodal metric learning aims to transform heterogeneous data into a common subspace where cross-modal similarity computing can be directly performed and has received much attention in recent years. Typically, the existing methods are designed for nonhierarchical labeled data. Such methods fail to exploit the intercategory correlations in the label hierarchy and, therefore, cannot achieve optimal performance on hierarchical labeled data. To address this problem, we propose a novel metric learning method for hierarchical labeled multimodal data, named deep hierarchical multimodal metric learning (DHMML). It learns the multilayer representations for each modality by establishing a layer-specific network corresponding to each layer in the label hierarchy. In particular, a multilayer classification mechanism is introduced to enable the layerwise representations to not only preserve the semantic similarities within each layer, but also retain the intercategory correlations across different layers. In addition, an adversarial learning mechanism is proposed to bridge the cross-modality gap by producing indistinguishable features for different modalities. Through integration of the multilayer classification and adversarial learning mechanisms, DHMML can obtain hierarchical discriminative modality-invariant representations for multimodal data. Experiments on two benchmark datasets are used to demonstrate the superiority of the proposed DHMML method over several state-of-the-art methods.

Visible-infrared person re-identification via specific and shared representations learning

The primary goal of visible-infrared person re-identification (VI-ReID) is to match pedestrian photos obtained during the day and night. The majority of existing methods simply generate auxiliary modalities to reduce the modality discrepancy for cross-modality matching. They capture modality-invariant representations but ignore the extraction of modality-specific representations that can aid in distinguishing among various identities of the same modality. To alleviate these issues, this work provides a novel specific and shared representations learning (SSRL) model for VI-ReID to learn modality-specific and modality-shared representations. We design a shared branch in SSRL to bridge the image-level gap and learn modality-shared representations, while a specific branch retains the discriminative information of visible images to learn modality-specific representations. In addition, we propose intra-class aggregation and inter-class separation learning strategies to optimize the distribution of feature embeddings at a fine-grained level. Extensive experimental results on two challenging benchmark datasets, SYSU-MM01 and RegDB, demonstrate the superior performance of SSRL over state-of-the-art methods.

Dual-Stream Transformer With Distribution Alignment for Visible-Infrared Person Re-Identification

Visible-infrared person re-identification(VI-ReID) aims to match the person images captured by visible and infrared cameras and suffers from severe cross-modality discrepancy and intra-modality variations. Existing approaches mainly use convolution neural network (CNN)-based architectures to extract pedestrian features, which fail to capture the long-range dependencies within an image. In addition, previous works usually attempt to bridge the modality gap by using adversarial learning to generate style-consistent images or designing different feature-level metric learning constraints. However, few works consider the cross-modality disparity from the perspective of assessing overall distance distribution discrepancy. To address these problems, we design a pure Transformer-based Visible-Infrared (TransVI) network with a conventional two-stream structure, which can explicitly capture modality-specific representations and learn multi-modality sharable knowledge. TransVI can efficiently address the lack of global dependency in CNN-based architectures due to the multi-head self-attention modules in the transformer, which allows us to capture the long-range dependencies of pedestrian images. Furthermore, we introduce the Cross-Modality Dissimilarity-based Maximum Mean Discrepancy (CMD-MMD) constraint to handle the cross-modality discrepancy at the distance distribution level. Specifically, CMD-MMD leverages intra-modality distribution separability to guide inter-modality distribution separability learning, aligning pair-wise distance distributions of intra- and inter-modality for within-class and between-class, respectively. In this way, the distance distributions of intra- and inter-modality become more similar, significantly mitigating the cross-modality discrepancy and learning more modality invariant representations. Extensive experimental results on two public VI-ReID datasets confirm that our proposed framework can achieve state-of-the-art performance.

Leveraging Modality-Specific Representations for Audio-Visual Speech Recognition via Reinforcement Learning

Audio-visual speech recognition (AVSR) has gained remarkable success for ameliorating the noise-robustness of speech recognition. Mainstream methods focus on fusing audio and visual inputs to obtain modality-invariant representations. However, such representations are prone to over-reliance on audio modality as it is much easier to recognize than video modality in clean conditions. As a result, the AVSR model underestimates the importance of visual stream in face of noise corruption. To this end, we leverage visual modality-specific representations to provide stable complementary information for the AVSR task. Specifically, we propose a reinforcement learning (RL) based framework called MSRL, where the agent dynamically harmonizes modality-invariant and modality-specific representations in the auto-regressive decoding process. We customize a reward function directly related to task-specific metrics (i.e., word error rate), which encourages the MSRL to effectively explore the optimal integration strategy. Experimental results on the LRS3 dataset show that the proposed method achieves state-of-the-art in both clean and various noisy conditions. Furthermore, we demonstrate the better generality of MSRL system than other baselines when test set contains unseen noises.

A modality-collaborative convolution and transformer hybrid network for unpaired multi-modal medical image segmentation with limited annotations.

Multi-modal learning is widely adopted to learn the latent complementary information between different modalities in multi-modal medical image segmentation tasks. Nevertheless, the traditional multi-modal learning methods require spatially well-aligned and paired multi-modal images for supervised training, which cannot leverage unpaired multi-modal images with spatial misalignment and modality discrepancy. For training accurate multi-modal segmentation networks using easily accessible and low-cost unpaired multi-modal images in clinical practice, unpaired multi-modal learning has received comprehensive attention recently. Existing unpaired multi-modal learning methods usually focus on the intensity distribution gap but ignore the scale variation problem between different modalities. Besides, within existing methods, shared convolutional kernels are frequently employed to capture common patterns in all modalities, but they are typically inefficient at learning global contextual information. On the other hand, existing methods highly rely on a large number of labeled unpaired multi-modal scans for training, which ignores the practical scenario when labeled data is limited. To solve the above problems, we propose a modality-collaborative convolution and transformer hybrid network (MCTHNet) using semi-supervised learning for unpaired multi-modal segmentation with limited annotations, which not only collaboratively learns modality-specific and modality-invariant representations, but also could automatically leverage extensive unlabeled scans for improving performance. We make three main contributions to the proposed method. First, to alleviate the intensity distribution gap and scale variation problems across modalities, we develop a modality-specific scale-aware convolution (MSSC) module that can adaptively adjust the receptive field sizes and feature normalization parameters according to the input. Secondly, we propose a modality-invariant vision transformer (MIViT) module as the shared bottleneck layer for all modalities, which implicitly incorporates convolution-like local operations with the global processing of transformers for learning generalizable modality-invariant representations. Third, we design a multi-modal cross pseudo supervision (MCPS) method for semi-supervised learning, which enforces the consistency between the pseudo segmentation maps generated by two perturbed networks to acquire abundant annotation information from unlabeled unpaired multi-modal scans. Extensive experiments are performed on two unpaired CT and MR segmentation datasets, including a cardiac substructure dataset derived from the MMWHS-2017 dataset and an abdominal multi-organ dataset consisting of the BTCV and CHAOS datasets. Experiment results show that our proposed method significantly outperforms other existing state-of-the-art methods under various labeling ratios, and achieves a comparable segmentation performance close to single-modal methods with fully labeled data by only leveraging a small portion of labeled data. Specifically, when the labeling ratio is 25%, our proposed method achieves overall mean DSC values of 78.56% and 76.18% in cardiac and abdominal segmentation, respectively, which significantly improves the average DSC value of two tasks by 12.84% compared to single-modal U-Net models. Our proposed method is beneficial for reducing the annotation burden of unpaired multi-modal medical images in clinical applications.

CAMR: cross-aligned multimodal representation learning for cancer survival prediction.

Accurately predicting cancer survival is crucial for helping clinicians to plan appropriate treatments, which largely improves the life quality of cancer patients and spares the related medical costs. Recent advances in survival prediction methods suggest that integrating complementary information from different modalities, e.g. histopathological images and genomic data, plays a key role in enhancing predictive performance. Despite promising results obtained by existing multimodal methods, the disparate and heterogeneous characteristics of multimodal data cause the so-called modality gap problem, which brings in dramatically diverse modality representations in feature space. Consequently, detrimental modality gaps make it difficult for comprehensive integration of multimodal information via representation learning and therefore pose a great challenge to further improvements of cancer survival prediction. To solve the above problems, we propose a novel method called cross-aligned multimodal representation learning (CAMR), which generates both modality-invariant and -specific representations for more accurate cancer survival prediction. Specifically, a cross-modality representation alignment learning network is introduced to reduce modality gaps by effectively learning modality-invariant representations in a common subspace, which is achieved by aligning the distributions of different modality representations through adversarial training. Besides, we adopt a cross-modality fusion module to fuse modality-invariant representations into a unified cross-modality representation for each patient. Meanwhile, CAMR learns modality-specific representations which complement modality-invariant representations and therefore provides a holistic view of the multimodal data for cancer survival prediction. Comprehensive experiment results demonstrate that CAMR can successfully narrow modality gaps and consistently yields better performance than other survival prediction methods using multimodal data. CAMR is freely available at https://github.com/wxq-ustc/CAMR. Supplementary data are available at Bioinformatics online.

Self-Supervised Correlation Learning for Cross-Modal Retrieval

Cross-modal retrieval aims to retrieve relevant data from another modality when given a query of one modality. Although most existing methods that rely on the label information of multimedia data have achieved promising results, the performance benefiting from labeled data comes at a high cost since labeling data often requires enormous labor resources, especially on large-scale multimedia datasets. Therefore, unsupervised cross-modal learning is of crucial importance in real-world applications. In this paper, we propose a novel unsupervised cross-modal retrieval method, named Self-supervised Correlation Learning (SCL), which takes full advantage of large amounts of unlabeled data to learn discriminative and modality-invariant representations. Since unsupervised learning lacks the supervision of category labels, we incorporate the knowledge from the input as a supervisory signal by maximizing the mutual information between the input and the output of different modality-specific projectors. Besides, for the purpose of learning discriminative representations, we exploit unsupervised contrastive learning to model the relationship among intra- and inter-modality instances, which makes similar samples closer and pushes dissimilar samples apart. Moreover, to further eliminate the modality gap, we use a weight-sharing scheme and minimize the modality-invariant loss in the joint representation space. Beyond that, we also extend the proposed method to the semi-supervised setting. Extensive experiments conducted on three widely-used benchmark datasets demonstrate that our method achieves competitive results compared with current state-of-the-art cross-modal retrieval approaches.

Visible-Infrared Person Re-Identification via Cross-Modality Interaction Transformer

Visible-infrared person re-identification (VI Re-ID) is designed to match person images of the same identity from visible and infrared cameras. Transformer structures have been successfully applied in the field of VI Re-ID. However, previous Transformer-based methods were mainly designed to capture global content information in a single modality, and could not simultaneously perceive semantic information between two modalities from a global perspective. To solve this problem, we propose a novel framework named the cross-modality interaction Transformer (CMIT). It has strong abilities in modeling spatial and sequential features that can capture dependencies between long-range features, and explicitly improves the discriminativeness of features by exchanging information across modalities, thus contributing to obtaining modality-invariant representations. Specifically, CMIT utilizes a cross-modality attention mechanism to enrich the feature representations of each patch token by interacting with the patch tokens of the other modality, and aggregates local features of the CNN structure and global information of the Transformer structure to mine feature saliency representation. Furthermore, the modality-discriminative (MD) loss function is proposed to learn potential consistency between modalities to encourage intra-modality compactness within class and inter-modality separation between classes. Extensive experiments on two benchmarks demonstrate that our approach outperforms state-of-the-art methods.

Modality-invariant temporal representation learning for multimodal sentiment classification

Multimodal sentiment classification is a notable research field that aims to refine sentimental information and classify the sentiment tendency from sequential multimodal data. Most existing sentimental recognition algorithms explore multimodal fusion schemes that achieve good performance. However, there are two key challenges to overcome. First, it is essential to effectively extract inter- and intra-modality features prior to fusion, while simultaneously reducing ambiguity. The second challenge is how to learn modality-invariant representations that capture the underlying similarities. In this paper, we present a modality-invariant temporal learning technique and a new gated inter-modality attention mechanism to overcome these issues. For the first challenge, our proposed gated inter-modality attention mechanism performs modality interactions and filters inconsistencies from multiple modalities in an adaptive manner. We also use parallel structures to learn more comprehensive sentimental information in pairs (i.e., acoustic and visual). In addition, to address the second problem, we treat each modality as a multivariate Gaussian distribution (considering each timestamp as a single Gaussian distribution) and use the KL divergence to capture the implicit temporal distribution-level similarities. These strategies are helpful in reducing domain shifts between different modalities and extracting effective sequential modality-invariant representations. We have conducted experiments on several public datasets (i.e., YouTube and MOUD) and the results show that our proposed method outperforms the state-of-the-art multimodal sentiment categorization methods.

Adversarial Decoupling and Modality-Invariant Representation Learning for Visible-Infrared Person Re-Identification

Visible-infrared person re-identification (RGB-IR ReID) has now attracted increasing attention due to its surveillance applications under low-light environments. However, the large intra-class variations between different domains are still a challenging issue in the field of computer vision. To address the above issue, we propose a novel adversarial Decoupling and Modality-invariant Representation learning (DMiR) method to explore potential spectrum-invariant yet identity-discriminative representations for cross-modality pedestrians. Our model consists of three key components, including Domain-related Representation Disentanglement (DrRD), Modality-invariant Discriminative Representation (MiDR) and Representation Orthogonal Decorrelation (ROD). First, two subnets named Identity-Net and Domain-Net are designed to extract identity-related features and domain-related features, respectively. Given this two-stream structure, the DrRD is introduced to achieve adversarial decoupling against domain-specific features via a min-max disentanglement process. Specifically, the classification objective function on Domain-Net is minimized to extract spectrum-specific information while maximizing it to reduce domain-specific information. Second, in Identity-Net, we introduce MiDR to enhance intra-class compactness and reduce domain variations by exploring positive and negative pair variations, semantic-wise differences, and pair-wise semantic variations. Finally, the correlation between the two decomposed features, i.e., identity-related features and domain-related features, may lead to the introduction of modal information in identity representations, and vice versa. Therefore, we present the ROD constraint to make the two decomposed features unrelated to each other, which can more effectively separate the two-component features and enhance feature representations. Practically, we construct ROD at the feature-level and parameter-level, and finally select feature-level ROD as the decorrelation strategy because of its superior decorrelation performance. The whole scheme leads to disentangling spectrum-dependent information, as well as purifying identity information. Extensive experiments are carried out on two mainstream RGB-IR ReID datasets, and the results demonstrate the effectiveness of our method.

Semi-supervised cross-modal retrieval with graph-based semantic alignment network

Semi-supervised cross-modal retrieval is an eclectic paradigm which learns common representations via exploiting underlying semantic information from both labeled and unlabeled data. Most existing methods ignore the rich semantic information of text data and are unable to fully utilize the text data in common representation learning. Moreover, they only considered the correlation of the data with the same semantic label, but ignored the correlation between the data with different semantic label. In this paper, we propose a novel semi-supervised cross-modal retrieval method, called Graph-based Semantic Alignment Network (GSAN), which learns common representation by aligning the features of different modalities with semantic embeddings of text data. Firstly, we design a Deep Supervised Semantic Encoding (DSSE) module to train the semantic projector and label predictor which can exploit the semantic embeddings and the predicted labels from unlabeled data of text modality. Then, GAN-based Bidirectional Fusion (GBF) module is designed to learn the mapping networks of two modalities (image and text). In order to make the mapping networks generate semantically discriminative and modality-invariant representations, we utilize the underlying semantic information exploited by DSSE to construct Graph-based Triplet Constraint (GTC) which can enforce feature embeddings from the semantically-matched (image and text) pairs to be more similar and push those mismatched ones away. By the benefit of fully using of semantic information, our approach can only use fewer label data and achieves the performance of state-of-the-art methods. In addition, since we only utilize the mapping networks trained in GBF module to generate common representations in referring stage, our approach is efficient and time saving in real world application. Extensive experiments on four widely-used datasets show the effectiveness of GSAN.

DA-GAN: Dual Attention Generative Adversarial Network for Cross-Modal Retrieval

Cross-modal retrieval aims to search samples of one modality via queries of other modalities, which is a hot issue in the community of multimedia. However, two main challenges, i.e., heterogeneity gap and semantic interaction across different modalities, have not been solved efficaciously. Reducing the heterogeneous gap can improve the cross-modal similarity measurement. Meanwhile, modeling cross-modal semantic interaction can capture the semantic correlations more accurately. To this end, this paper presents a novel end-to-end framework, called Dual Attention Generative Adversarial Network (DA-GAN). This technique is an adversarial semantic representation model with a dual attention mechanism, i.e., intra-modal attention and inter-modal attention. Intra-modal attention is used to focus on the important semantic feature within a modality, while inter-modal attention is to explore the semantic interaction between different modalities and then represent the high-level semantic correlation more precisely. A dual adversarial learning strategy is designed to generate modality-invariant representations, which can reduce the cross-modal heterogeneity efficiently. The experiments on three commonly used benchmarks show the better performance of DA-GAN than these competitors.

An Intrinsic Structured Graph Alignment Module With Modality-Invariant Representations for NIR-VIS Face Recognition

Most existing near-infrared to visible (NIR-VIS) face recognition (FR) methods rely on global feature representations to reduce cross-modality discrepancies, but ignore the structural relationships between local features, e.g., the relative positions of eyes, nose, and mouth. Precise alignment of these local features can enhance the learning of modality-invariant face representations, thereby improving the performance of NIR-VIS FR. Therefore, in this letter, we propose an intrinsic structured graph alignment (ISGA) module that aims to obtain a graph-level local feature alignment across modalities. To this end, we first construct an intrinsic structure graph to model the inherent structural relationships of local features, and then enhance the discriminative feature representation by aligning the graphs between modalities. To jointly encourage cross-modality class consistency between semantics and structural relationships, a cross-modality class distribution (CMCD) loss is proposed to add an identity-preserving constraint to each class distribution in the embedding space between two modalities. To solve the resulting problem of suppressed class divisibility, we maximize the mutual information between inputs and class predictions. Extensive experiments on challenging NIR-VIS datasets indicate that our approach outperforms the state-of-the-arts. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/JianYu777/ISGA-CMCD</uri> .

Integrating Multi-Label Contrastive Learning With Dual Adversarial Graph Neural Networks for Cross-Modal Retrieval.

With the growing amount of multimodal data, cross-modal retrieval has attracted more and more attention and become a hot research topic. To date, most of the existing techniques mainly convert multimodal data into a common representation space where similarities in semantics between samples can be easily measured across multiple modalities. However, these approaches may suffer from the following limitations: 1) They overcome the modality gap by introducing loss in the common representation space, which may not be sufficient to eliminate the heterogeneity of various modalities; 2) They treat labels as independent entities and ignore label relationships, which is not conducive to establishing semantic connections across multimodal data; 3) They ignore the non-binary values of label similarity in multi-label scenarios, which may lead to inefficient alignment of representation similarity with label similarity. To tackle these problems, in this article, we propose two models to learn discriminative and modality-invariant representations for cross-modal retrieval. First, the dual generative adversarial networks are built to project multimodal data into a common representation space. Second, to model label relation dependencies and develop inter-dependent classifiers, we employ multi-hop graph neural networks (consisting of Probabilistic GNN and Iterative GNN), where the layer aggregation mechanism is suggested for using propagation information of various hops. Third, we propose a novel soft multi-label contrastive loss for cross-modal retrieval, with the soft positive sampling probability, which can align the representation similarity and the label similarity. Additionally, to adapt to incomplete-modal learning, which can have wider applications, we propose a modal reconstruction mechanism to generate missing features. Extensive experiments on three widely used benchmark datasets, i.e., NUS-WIDE, MIRFlickr, and MS-COCO, show the superiority of our proposed method.

Linear Cross-Modal Hash Encoding (LCMHE) for Visual and Near-Infrared Face Recognition

Face recognition in practical situations often encounters the change of acquisition sensors, i.e., the gallery images are captured in near-infrared and probe images captured in the visual domain or vice versa. Thus, for robust face recognition under such circumstance, a modality invariant representation is required. This letter proposes a hash-encoding-based descriptor to address this modality gap for face recognition problem, namely linear cross-modal hash encoding (LCMHE). LCMHE encodes the pixels of the face images based on a pretrained hash dictionary. The encoding highlights the reflectance part of the face and enables accurate recognition. We further explored the possibility of LCMHE by combining the deep-learning model to get better recognition. We experimented with publicly available CASIA NIR-VIS 2.0 databases and verified the effectiveness of our proposed letter.

Deep learning based feature-level integration of multi-omics data for breast cancer patients survival analysis

BackgroundBreast cancer is the most prevalent and among the most deadly cancers in females. Patients with breast cancer have highly variable survival lengths, indicating a need to identify prognostic biomarkers for personalized diagnosis and treatment. With the development of new technologies such as next-generation sequencing, multi-omics information are becoming available for a more thorough evaluation of a patient’s condition. In this study, we aim to improve breast cancer overall survival prediction by integrating multi-omics data (e.g., gene expression, DNA methylation, miRNA expression, and copy number variations (CNVs)).MethodsMotivated by multi-view learning, we propose a novel strategy to integrate multi-omics data for breast cancer survival prediction by applying complementary and consensus principles. The complementary principle assumes each -omics data contains modality-unique information. To preserve such information, we develop a concatenation autoencoder (ConcatAE) that concatenates the hidden features learned from each modality for integration. The consensus principle assumes that the disagreements among modalities upper bound the model errors. To get rid of the noises or discrepancies among modalities, we develop a cross-modality autoencoder (CrossAE) to maximize the agreement among modalities to achieve a modality-invariant representation. We first validate the effectiveness of our proposed models on the MNIST simulated data. We then apply these models to the TCCA breast cancer multi-omics data for overall survival prediction.ResultsFor breast cancer overall survival prediction, the integration of DNA methylation and miRNA expression achieves the best overall performance of 0.641 ± 0.031 with ConcatAE, and 0.63 ± 0.081 with CrossAE. Both strategies outperform baseline single-modality models using only DNA methylation (0.583 ± 0.058) or miRNA expression (0.616 ± 0.057).ConclusionsIn conclusion, we achieve improved overall survival prediction performance by utilizing either the complementary or consensus information among multi-omics data. The proposed ConcatAE and CrossAE models can inspire future deep representation-based multi-omics integration techniques. We believe these novel multi-omics integration models can benefit the personalized diagnosis and treatment of breast cancer patients.

Semi-Supervised Cross-Modal Retrieval With Label Prediction

Due to abundance of data from multiple modalities, cross-modal retrieval tasks with image-text, audio-image, etc. are gaining increasing importance. Of the different approaches proposed, supervised methods usually give significant improvement over their unsupervised counterparts at the additional cost of labeling or annotation of the training data. Semi-supervised methods are recently becoming popular as they provide an elegant framework to balance the conflicting requirement of labeling cost and accuracy. In this work, we propose a novel deep semi-supervised framework which can seamlessly handle both labeled as well as unlabeled data. The network has two important components: (a) the label prediction component predicts the labels for the unlabeled portion of the data and then (b) a common modality-invariant representation is learned for cross-modal retrieval. The two parts of the network are trained sequentially one after the other. Extensive experiments on three standard benchmark datasets, Wiki, Pascal VOC and NUS-WIDE demonstrate that the proposed framework outperforms the state-of-the-art for both supervised and semi-supervised settings.