Huiguang He

Changde Du, Kaicheng Fu, Jinpeng Li et al.

Decoding human visual neural representations is a challenging task with great scientific significance in revealing vision-processing mechanisms and developing brain-like intelligent machines. Most existing methods are difficult to generalize to novel categories that have no corresponding neural data for training. The two main reasons are 1) the under-exploitation of the multimodal semantic knowledge underlying the neural data and 2) the small number of paired (stimuli-responses) training data. To overcome these limitations, this paper presents a generic neural decoding method called BraVL that uses multimodal learning of brain-visual-linguistic features. We focus on modeling the relationships between brain, visual and linguistic features via multimodal deep generative models. Specifically, we leverage the mixture-of-product-of-experts formulation to infer a latent code that enables a coherent joint generation of all three modalities. To learn a more consistent joint representation and improve the data efficiency in the case of limited brain activity data, we exploit both intra- and inter-modality mutual information maximization regularization terms. In particular, our BraVL model can be trained under various semi-supervised scenarios to incorporate the visual and textual features obtained from the extra categories. Finally, we construct three trimodal matching datasets, and the extensive experiments lead to some interesting conclusions and cognitive insights: 1) decoding novel visual categories from human brain activity is practically possible with good accuracy; 2) decoding models using the combination of visual and linguistic features perform much better than those using either of them alone; 3) visual perception may be accompanied by linguistic influences to represent the semantics of visual stimuli. Code and data: https://github.com/ChangdeDu/BraVL.

Yizhuo Lu, Changde Du, Qiongyi zhou et al.

Reconstructing visual stimuli from brain recordings has been a meaningful and challenging task. Especially, the achievement of precise and controllable image reconstruction bears great significance in propelling the progress and utilization of brain-computer interfaces. Despite the advancements in complex image reconstruction techniques, the challenge persists in achieving a cohesive alignment of both semantic (concepts and objects) and structure (position, orientation, and size) with the image stimuli. To address the aforementioned issue, we propose a two-stage image reconstruction model called MindDiffuser. In Stage 1, the VQ-VAE latent representations and the CLIP text embeddings decoded from fMRI are put into Stable Diffusion, which yields a preliminary image that contains semantic information. In Stage 2, we utilize the CLIP visual feature decoded from fMRI as supervisory information, and continually adjust the two feature vectors decoded in Stage 1 through backpropagation to align the structural information. The results of both qualitative and quantitative analyses demonstrate that our model has surpassed the current state-of-the-art models on Natural Scenes Dataset (NSD). The subsequent experimental findings corroborate the neurobiological plausibility of the model, as evidenced by the interpretability of the multimodal feature employed, which align with the corresponding brain responses.

Yizhuo Lu, Changde Du, Dianpeng Wang et al.

Reconstructing visual stimuli from measured functional magnetic resonance imaging (fMRI) has been a meaningful and challenging task. Previous studies have successfully achieved reconstructions with structures similar to the original images, such as the outlines and size of some natural images. However, these reconstructions lack explicit semantic information and are difficult to discern. In recent years, many studies have utilized multi-modal pre-trained models with stronger generative capabilities to reconstruct images that are semantically similar to the original ones. However, these images have uncontrollable structural information such as position and orientation. To address both of the aforementioned issues simultaneously, we propose a two-stage image reconstruction model called MindDiffuser, utilizing Stable Diffusion. In Stage 1, the VQ-VAE latent representations and the CLIP text embeddings decoded from fMRI are put into the image-to-image process of Stable Diffusion, which yields a preliminary image that contains semantic and structural information. In Stage 2, we utilize the low-level CLIP visual features decoded from fMRI as supervisory information, and continually adjust the two features in Stage 1 through backpropagation to align the structural information. The results of both qualitative and quantitative analyses demonstrate that our proposed model has surpassed the current state-of-the-art models in terms of reconstruction results on Natural Scenes Dataset (NSD). Furthermore, the results of ablation experiments indicate that each component of our model is effective for image reconstruction.

Zhongyu Huang, Yingheng Wang, Chaozhuo Li et al.

The invariance to permutations of the adjacency matrix, i.e., graph isomorphism, is an overarching requirement for Graph Neural Networks (GNNs). Conventionally, this prerequisite can be satisfied by the invariant operations over node permutations when aggregating messages. However, such an invariant manner may ignore the relationships among neighboring nodes, thereby hindering the expressivity of GNNs. In this work, we devise an efficient permutation-sensitive aggregation mechanism via permutation groups, capturing pairwise correlations between neighboring nodes. We prove that our approach is strictly more powerful than the 2-dimensional Weisfeiler-Lehman (2-WL) graph isomorphism test and not less powerful than the 3-WL test. Moreover, we prove that our approach achieves the linear sampling complexity. Comprehensive experiments on multiple synthetic and real-world datasets demonstrate the superiority of our model.

Zhongying Deng, Cheng Tang, Ziyan Huang et al. · pku

Foundation models have demonstrated remarkable success across diverse domains and tasks, primarily due to the thrive of large-scale, diverse, and high-quality datasets. However, in the field of medical imaging, the curation and assembling of such medical datasets are highly challenging due to the reliance on clinical expertise and strict ethical and privacy constraints, resulting in a scarcity of large-scale unified medical datasets and hindering the development of powerful medical foundation models. In this work, we present the largest survey to date of medical image datasets, covering over 1,000 open-access datasets with a systematic catalog of their modalities, tasks, anatomies, annotations, limitations, and potential for integration. Our analysis exposes a landscape that is modest in scale, fragmented across narrowly scoped tasks, and unevenly distributed across organs and modalities, which in turn limits the utility of existing medical image datasets for developing versatile and robust medical foundation models. To turn fragmentation into scale, we propose a metadata-driven fusion paradigm (MDFP) that integrates public datasets with shared modalities or tasks, thereby transforming multiple small data silos into larger, more coherent resources. Building on MDFP, we release an interactive discovery portal that enables end-to-end, automated medical image dataset integration, and compile all surveyed datasets into a unified, structured table that clearly summarizes their key characteristics and provides reference links, offering the community an accessible and comprehensive repository. By charting the current terrain and offering a principled path to dataset consolidation, our survey provides a practical roadmap for scaling medical imaging corpora, supporting faster data discovery, more principled dataset creation, and more capable medical foundation models.

Haoyu Lu, Qiongyi Zhou, Nanyi Fei et al.

Multimodal learning, especially large-scale multimodal pre-training, has developed rapidly over the past few years and led to the greatest advances in artificial intelligence (AI). Despite its effectiveness, understanding the underlying mechanism of multimodal pre-training models still remains a grand challenge. Revealing the explainability of such models is likely to enable breakthroughs of novel learning paradigms in the AI field. To this end, given the multimodal nature of the human brain, we propose to explore the explainability of multimodal learning models with the aid of non-invasive brain imaging technologies such as functional magnetic resonance imaging (fMRI). Concretely, we first present a newly-designed multimodal foundation model pre-trained on 15 million image-text pairs, which has shown strong multimodal understanding and generalization abilities in a variety of cognitive downstream tasks. Further, from the perspective of neural encoding (based on our foundation model), we find that both visual and lingual encoders trained multimodally are more brain-like compared with unimodal ones. Particularly, we identify a number of brain regions where multimodally-trained encoders demonstrate better neural encoding performance. This is consistent with the findings in existing studies on exploring brain multi-sensory integration. Therefore, we believe that multimodal foundation models are more suitable tools for neuroscientists to study the multimodal signal processing mechanisms in the human brain. Our findings also demonstrate the potential of multimodal foundation models as ideal computational simulators to promote both AI-for-brain and brain-for-AI research.

Kaicheng Fu, Changde Du, Shengpei Wang et al.

Decoding emotional states from human brain activity plays an important role in brain-computer interfaces. Existing emotion decoding methods still have two main limitations: one is only decoding a single emotion category from a brain activity pattern and the decoded emotion categories are coarse-grained, which is inconsistent with the complex emotional expression of human; the other is ignoring the discrepancy of emotion expression between the left and right hemispheres of human brain. In this paper, we propose a novel multi-view multi-label hybrid model for fine-grained emotion decoding (up to 80 emotion categories) which can learn the expressive neural representations and predicting multiple emotional states simultaneously. Specifically, the generative component of our hybrid model is parametrized by a multi-view variational auto-encoder, in which we regard the brain activity of left and right hemispheres and their difference as three distinct views, and use the product of expert mechanism in its inference network. The discriminative component of our hybrid model is implemented by a multi-label classification network with an asymmetric focal loss. For more accurate emotion decoding, we first adopt a label-aware module for emotion-specific neural representations learning and then model the dependency of emotional states by a masked self-attention mechanism. Extensive experiments on two visually evoked emotional datasets show the superiority of our method.

Changde Du, Kaicheng Fu, Bincheng Wen et al.

Understanding how humans conceptualize and categorize natural objects offers critical insights into perception and cognition. With the advent of Large Language Models (LLMs), a key question arises: can these models develop human-like object representations from linguistic and multimodal data? In this study, we combined behavioral and neuroimaging analyses to explore the relationship between object concept representations in LLMs and human cognition. We collected 4.7 million triplet judgments from LLMs and Multimodal LLMs (MLLMs) to derive low-dimensional embeddings that capture the similarity structure of 1,854 natural objects. The resulting 66-dimensional embeddings were stable, predictive, and exhibited semantic clustering similar to human mental representations. Remarkably, the dimensions underlying these embeddings were interpretable, suggesting that LLMs and MLLMs develop human-like conceptual representations of objects. Further analysis showed strong alignment between model embeddings and neural activity patterns in brain regions such as EBA, PPA, RSC, and FFA. This provides compelling evidence that the object representations in LLMs, while not identical to human ones, share fundamental similarities that reflect key aspects of human conceptual knowledge. Our findings advance the understanding of machine intelligence and inform the development of more human-like artificial cognitive systems.

Zhongyu Huang, Changde Du, Chaozhuo Li et al.

The brain basis of emotion has consistently received widespread attention, attracting a large number of studies to explore this cutting-edge topic. However, the methods employed in these studies typically only model the pairwise relationship between two brain regions, while neglecting the interactions and information fusion among multiple brain regions$\unicode{x2014}$one of the key ideas of the psychological constructionist hypothesis. To overcome the limitations of traditional methods, this study provides an in-depth theoretical analysis of how to maximize interactions and information fusion among brain regions. Building on the results of this analysis, we propose to identify the hierarchical emotional areas in the human brain through multi-source information fusion and graph machine learning methods. Comprehensive experiments reveal that the identified hierarchical emotional areas, from lower to higher levels, primarily facilitate the fundamental process of emotion perception, the construction of basic psychological operations, and the coordination and integration of these operations. Overall, our findings provide unique insights into the brain mechanisms underlying specific emotions based on the psychological constructionist hypothesis.

Shuangchen Zhao, Changde Du, Hui Li et al.

Deep Neural Networks (DNNs) have demonstrated exceptional recognition capabilities in traditional computer vision (CV) tasks. However, existing CV models often suffer a significant decrease in accuracy when confronted with out-of-distribution (OOD) data. In contrast to these DNN models, human can maintain a consistently low error rate when facing OOD scenes, partly attributed to the rich prior cognitive knowledge stored in the human brain. Previous OOD generalization researches only focus on the single modal, overlooking the advantages of multimodal learning method. In this paper, we utilize the multimodal learning method to improve the OOD generalization and propose a novel Brain-machine Fusion Learning (BMFL) framework. We adopt the cross-attention mechanism to fuse the visual knowledge from CV model and prior cognitive knowledge from the human brain. Specially, we employ a pre-trained visual neural encoding model to predict the functional Magnetic Resonance Imaging (fMRI) from visual features which eliminates the need for the fMRI data collection and pre-processing, effectively reduces the workload associated with conventional BMFL methods. Furthermore, we construct a brain transformer to facilitate the extraction of knowledge inside the fMRI data. Moreover, we introduce the Pearson correlation coefficient maximization regularization method into the training process, which improves the fusion capability with better constrains. Our model outperforms the DINOv2 and baseline models on the ImageNet-1k validation dataset as well as six curated OOD datasets, showcasing its superior performance in diverse scenarios.

Qiongyi Zhou, Changde Du, Shengpei Wang et al.

The study of decoding visual neural information faces challenges in generalizing single-subject decoding models to multiple subjects, due to individual differences. Moreover, the limited availability of data from a single subject has a constraining impact on model performance. Although prior multi-subject decoding methods have made significant progress, they still suffer from several limitations, including difficulty in extracting global neural response features, linear scaling of model parameters with the number of subjects, and inadequate characterization of the relationship between neural responses of different subjects to various stimuli. To overcome these limitations, we propose a CLIP-guided Multi-sUbject visual neural information SEmantic Decoding (CLIP-MUSED) method. Our method consists of a Transformer-based feature extractor to effectively model global neural representations. It also incorporates learnable subject-specific tokens that facilitates the aggregation of multi-subject data without a linear increase of parameters. Additionally, we employ representational similarity analysis (RSA) to guide token representation learning based on the topological relationship of visual stimuli in the representation space of CLIP, enabling full characterization of the relationship between neural responses of different subjects under different stimuli. Finally, token representations are used for multi-subject semantic decoding. Our proposed method outperforms single-subject decoding methods and achieves state-of-the-art performance among the existing multi-subject methods on two fMRI datasets. Visualization results provide insights into the effectiveness of our proposed method. Code is available at https://github.com/CLIP-MUSED/CLIP-MUSED.

Xiaoyu Chen, Changde Du, Che Liu et al.

Decoding language information from brain signals represents a vital research area within brain-computer interfaces, particularly in the context of deciphering the semantic information from the fMRI signal. However, many existing efforts concentrate on decoding small vocabulary sets, leaving space for the exploration of open vocabulary continuous text decoding. In this paper, we introduce a novel method, the \textbf{Brain Prompt GPT (BP-GPT)}. By using the brain representation that is extracted from the fMRI as a prompt, our method can utilize GPT-2 to decode fMRI signals into stimulus text. Further, we introduce a text-to-text baseline and align the fMRI prompt to the text prompt. By introducing the text-to-text baseline, our BP-GPT can extract a more robust brain prompt and promote the decoding of pre-trained LLM. We evaluate our BP-GPT on the open-source auditory semantic decoding dataset and achieve a significant improvement up to $4.61\%$ on METEOR and $2.43\%$ on BERTScore across all the subjects compared to the state-of-the-art method. The experimental results demonstrate that using brain representation as a prompt to further drive LLM for auditory neural decoding is feasible and effective.

Hui He, Kun Yi, Yuanchi Ma et al.

The recent boom of large pre-trained models witnesses remarkable success in developing foundation models (FMs) for time series forecasting. Despite impressive performance across diverse downstream forecasting tasks, existing time series FMs possess massive network architectures and require substantial pre-training on large-scale datasets, which significantly hinders their deployment in resource-constrained environments. In response to this growing tension between versatility and affordability, we propose SEMPO, a novel lightweight foundation model that requires pretraining on relatively small-scale data, yet exhibits strong general time series forecasting. Concretely, SEMPO comprises two key modules: 1) energy-aware SpEctral decomposition module, that substantially improves the utilization of pre-training data by modeling not only the high-energy frequency signals but also the low-energy yet informative frequency signals that are ignored in current methods; and 2) Mixture-of-PrOmpts enabled Transformer, that learns heterogeneous temporal patterns through small dataset-specific prompts and adaptively routes time series tokens to prompt-based experts for parameter-efficient model adaptation across different datasets and domains. Equipped with these modules, SEMPO significantly reduces both pre-training data scale and model size, while achieving strong generalization. Extensive experiments on two large-scale benchmarks covering 16 datasets demonstrate the superior performance of SEMPO in both zero-shot and few-shot forecasting scenarios compared with state-of-the-art methods. Code and data are available at https://github.com/mala-lab/SEMPO.

Hao Chen, Ming Jin, Zhunan Li et al.

As an essential element for the diagnosis and rehabilitation of psychiatric disorders, the electroencephalogram (EEG) based emotion recognition has achieved significant progress due to its high precision and reliability. However, one obstacle to practicality lies in the variability between subjects and sessions. Although several studies have adopted domain adaptation (DA) approaches to tackle this problem, most of them treat multiple EEG data from different subjects and sessions together as a single source domain for transfer, which either fails to satisfy the assumption of domain adaptation that the source has a certain marginal distribution, or increases the difficulty of adaptation. We therefore propose the multi-source marginal distribution adaptation (MS-MDA) for EEG emotion recognition, which takes both domain-invariant and domain-specific features into consideration. First, we assume that different EEG data share the same low-level features, then we construct independent branches for multiple EEG data source domains to adopt one-to-one domain adaptation and extract domain-specific features. Finally, the inference is made by multiple branches. We evaluate our method on SEED and SEED-IV for recognizing three and four emotions, respectively. Experimental results show that the MS-MDA outperforms the comparison methods and state-of-the-art models in cross-session and cross-subject transfer scenarios in our settings. Codes at https://github.com/VoiceBeer/MS-MDA.

Xujin Li, Wei Wei, Shuang Qiu et al.

The Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an efficient technology for target retrieval using electroencephalography (EEG) signals. The performance improvement of traditional decoding methods relies on a substantial amount of training data from new test subjects, which increases preparation time for BCI systems. Several studies introduce data from existing subjects to reduce the dependence of performance improvement on data from new subjects, but their optimization strategy based on adversarial learning with extensive data increases training time during the preparation procedure. Moreover, most previous methods only focus on the single-view information of EEG signals, but ignore the information from other views which may further improve performance. To enhance decoding performance while reducing preparation time, we propose a Temporal-Spectral fusion transformer with Subject-specific Adapter (TSformer-SA). Specifically, a cross-view interaction module is proposed to facilitate information transfer and extract common representations across two-view features extracted from EEG temporal signals and spectrogram images. Then, an attention-based fusion module fuses the features of two views to obtain comprehensive discriminative features for classification. Furthermore, a multi-view consistency loss is proposed to maximize the feature similarity between two views of the same EEG signal. Finally, we propose a subject-specific adapter to rapidly transfer the knowledge of the model trained on data from existing subjects to decode data from new subjects. Experimental results show that TSformer-SA significantly outperforms comparison methods and achieves outstanding performance with limited training data from new subjects. This facilitates efficient decoding and rapid deployment of BCI systems in practical use.

Huaxu He

Spiking Neural Networks (SNNs) are a class of network models capable of processing spatiotemporal information, with event-driven characteristics and energy efficiency advantages. Recently, directly trained SNNs have shown potential to match or surpass the performance of traditional Artificial Neural Networks (ANNs) in classification tasks. However, in object detection tasks, directly trained SNNs still exhibit a significant performance gap compared to ANNs when tested on frame-based static object datasets (such as COCO2017). Therefore, bridging this performance gap and enabling directly trained SNNs to achieve performance comparable to ANNs on these static datasets has become one of the key challenges in the development of SNNs.To address this challenge, this paper focuses on enhancing the SNN's unique ability to process spatiotemporal information. Spiking neurons, as the core components of SNNs, facilitate the exchange of information between different temporal channels during the process of converting input floating-point data into binary spike signals. However, existing neuron models still have certain limitations in the communication of temporal information. Some studies have even suggested that disabling the backpropagation in the time dimension during SNN training can still yield good training results. To improve the SNN handling of temporal information, this paper proposes replacing traditional 2D convolutions with 3D convolutions, thus directly incorporating temporal information into the convolutional process. Additionally, temporal information recurrence mechanism is introduced within the neurons to further enhance the neurons' efficiency in utilizing temporal information.Experimental results show that the proposed method enables directly trained SNNs to achieve performance levels comparable to ANNs on the COCO2017 and VOC datasets.

Penghua Zhai, Huaiwei Cong, Gangming Zhao et al.

\emph{Objective and Impact Statement}. With the renaissance of deep learning, automatic diagnostic systems for computed tomography (CT) have achieved many successful applications. However, they are mostly attributed to careful expert annotations, which are often scarce in practice. This drives our interest to the unsupervised representation learning. \emph{Introduction}. Recent studies have shown that self-supervised learning is an effective approach for learning representations, but most of them rely on the empirical design of transformations and pretext tasks. \emph{Methods}. To avoid the subjectivity associated with these methods, we propose the MVCNet, a novel unsupervised three dimensional (3D) representation learning method working in a transformation-free manner. We view each 3D lesion from different orientations to collect multiple two dimensional (2D) views. Then, an embedding function is learned by minimizing a contrastive loss so that the 2D views of the same 3D lesion are aggregated, and the 2D views of different lesions are separated. We evaluate the representations by training a simple classification head upon the embedding layer. \emph{Results}. Experimental results show that MVCNet achieves state-of-the-art accuracies on the LIDC-IDRI (89.55\%), LNDb (77.69\%) and TianChi (79.96\%) datasets for \emph{unsupervised representation learning}. When fine-tuned on 10\% of the labeled data, the accuracies are comparable to the supervised learning model (89.46\% vs. 85.03\%, 73.85\% vs. 73.44\%, 83.56\% vs. 83.34\% on the three datasets, respectively). \emph{Conclusion}. Results indicate the superiority of MVCNet in \emph{learning representations with limited annotations}.

Changde Du, Changying Du, Hao Wang et al.

There are threefold challenges in emotion recognition. First, it is difficult to recognize human's emotional states only considering a single modality. Second, it is expensive to manually annotate the emotional data. Third, emotional data often suffers from missing modalities due to unforeseeable sensor malfunction or configuration issues. In this paper, we address all these problems under a novel multi-view deep generative framework. Specifically, we propose to model the statistical relationships of multi-modality emotional data using multiple modality-specific generative networks with a shared latent space. By imposing a Gaussian mixture assumption on the posterior approximation of the shared latent variables, our framework can learn the joint deep representation from multiple modalities and evaluate the importance of each modality simultaneously. To solve the labeled-data-scarcity problem, we extend our multi-view model to semi-supervised learning scenario by casting the semi-supervised classification problem as a specialized missing data imputation task. To address the missing-modality problem, we further extend our semi-supervised multi-view model to deal with incomplete data, where a missing view is treated as a latent variable and integrated out during inference. This way, the proposed overall framework can utilize all available (both labeled and unlabeled, as well as both complete and incomplete) data to improve its generalization ability. The experiments conducted on two real multi-modal emotion datasets demonstrated the superiority of our framework.

Changde Du, Changying Du, Huiguang He

Decoding human brain activities via functional magnetic resonance imaging (fMRI) has gained increasing attention in recent years. While encouraging results have been reported in brain states classification tasks, reconstructing the details of human visual experience still remains difficult. Two main challenges that hinder the development of effective models are the perplexing fMRI measurement noise and the high dimensionality of limited data instances. Existing methods generally suffer from one or both of these issues and yield dissatisfactory results. In this paper, we tackle this problem by casting the reconstruction of visual stimulus as the Bayesian inference of missing view in a multiview latent variable model. Sharing a common latent representation, our joint generative model of external stimulus and brain response is not only "deep" in extracting nonlinear features from visual images, but also powerful in capturing correlations among voxel activities of fMRI recordings. The nonlinearity and deep structure endow our model with strong representation ability, while the correlations of voxel activities are critical for suppressing noise and improving prediction. We devise an efficient variational Bayesian method to infer the latent variables and the model parameters. To further improve the reconstruction accuracy, the latent representations of testing instances are enforced to be close to that of their neighbours from the training set via posterior regularization. Experiments on three fMRI recording datasets demonstrate that our approach can more accurately reconstruct visual stimuli.

Changde Du, Changying Du, Jinpeng Li et al.

In emotion recognition, it is difficult to recognize human's emotional states using just a single modality. Besides, the annotation of physiological emotional data is particularly expensive. These two aspects make the building of effective emotion recognition model challenging. In this paper, we first build a multi-view deep generative model to simulate the generative process of multi-modality emotional data. By imposing a mixture of Gaussians assumption on the posterior approximation of the latent variables, our model can learn the shared deep representation from multiple modalities. To solve the labeled-data-scarcity problem, we further extend our multi-view model to semi-supervised learning scenario by casting the semi-supervised classification problem as a specialized missing data imputation task. Our semi-supervised multi-view deep generative framework can leverage both labeled and unlabeled data from multiple modalities, where the weight factor for each modality can be learned automatically. Compared with previous emotion recognition methods, our method is more robust and flexible. The experiments conducted on two real multi-modal emotion datasets have demonstrated the superiority of our framework over a number of competitors.