Blood pressure, as a paramount and indispensable physiological health indicator for evaluating cardiovascular health, is often assessed using mainstream methods that combine multiple physiological signals in non-end-to-end measurement approaches. However, these methods are prone to introducing noise in low-level features during blood pressure waveform detection. To address this issue, we propose a U-Net-based multi-scale attention model architecture—BP-WaveNet. This network requires only PPG signals as input features and replaces the traditional encoder with a multi-scale attention mechanism, effectively suppressing noise in low-level features to a certain extent. Additionally, we designed and refined a residual connection structure, which combines residual paths with convolutional layers to enhance the network's ability to extract critical information such as heartbeat rhythm and blood flow acceleration. Experimental results demonstrate that BP-WaveNet achieves superior performance on the MIMIC-III dataset, with mean absolute errors (MAE) for systolic and diastolic blood pressure of 2.77 mmHg and 2.21 mmHg, respectively, meeting the standards set by the British Hypertension Society (BHS). Compared with other models, our method shows stronger generalization ability and lower error rate.

With the advancement of modern medicine and health monitoring technology, non-contact and non-invasive health monitoring methods have become a focal point of research and innovation. Blood pressure, as one of the most critical vital signs, plays a key role in individual health. Hypertension, as a major risk factor for cardiovascular diseases, poses a significant global health challenge. Traditional cuff-based blood pressure measurement methods are limited in providing real-time, convenient, and long-term monitoring, especially for the elderly and patients with chronic conditions.However, existing research on non-contact blood pressure prediction has limited practical applicability. For instance, blood pressure prediction based on facial videos requires stable lighting conditions and cannot function properly in environments with unstable or no light. This study proposes the MultiScale TransConv Network (MSTC-Net), which predicts blood pressure by fusing signals from different modalities.By fusing visible light and infrared light videos, MSTC-Net overcomes the limitations of traditional facial video-based methods that require stable lighting conditions. The method achieves high prediction accuracy on our collected ViInHealth dataset, with mean absolute errors (MAE) of 8.67 mmHg and 6.21 mmHg for systolic and diastolic blood pressure, respectively. The proposed technology offers a promising solution for non-invasive blood pressure monitoring, improving medical monitoring efficiency and contributing to the intelligent development of health management.

Genomic prediction (GP) uses single nucleotide polymorphisms (SNPs) to establish associations between markers and phenotypes. Selection of early individuals by genomic estimated breeding value shortens the generation interval and speeds up the breeding process. Recently, methods based on deep learning (DL) have gained great attention in the field of GP. In this study, we explore the application of Transformer-based structures to GP and develop a novel deep-learning model named GPformer. GPformer obtains a global view by gleaning beneficial information from all relevant SNPs regardless of the physical distance between SNPs. Comprehensive experimental results on five different crop datasets show that GPformer outperforms ridge regression-based linear unbiased prediction (RR-BLUP), support vector regression (SVR), light gradient boosting machine (LightGBM) and deep neural network genomic prediction (DNNGP) in terms of mean absolute error, Pearson's correlation coefficient and the proposed metric consistent index. Furthermore, we introduce a knowledge-guided module (KGM) to extract genome-wide association studies-based information, which is fused into GPformer as prior knowledge. KGM is very flexible and can be plugged into any DL network. Ablation studies of KGM on three datasets illustrate the efficiency of KGM adequately. Moreover, GPformer is robust and stable to hyperparameters and can generalize to each phenotype of every dataset, which is suitable for practical application scenarios.

Feature-map based knowledge distillation has exhibited its significance in improving the performance of student model. Existing works mainly focus on the formulation of knowledge, but ignore the number difference of channels due to heterogeneous architectures of teacher-student pair. They generally adopt handcrafted matching or input-independent association matrix, which would lead to the semantic mismatch, thus suboptimal performance. To resolve this problem, we present an input-dependent channel association module. This module automatically generates an allocation matrix in a cross-attention manner, which enables each student channel to be dynamically connected to its semantic-related teacher channel based on its learning state. An alternative training scheme is applied for stable optimization. Extensive experiments on image classification with a variety of settings based on the popular network architectures well demonstrate the effectiveness of our proposed strategy.

To robustly detect arbitrary-shaped scene texts, bottom-up methods are widely explored for their flexibility. Due to the highly homogeneous texture and cluttered distribution of scene texts, it is nontrivial for segmentation-based methods to discover the separatrixes between adjacent instances. To effectively separate nearby texts, many methods adopt the seed expansion strategy that segments shrunken text regions as seed areas, and then iteratively expands the seed areas into intact text regions. In seek of a more straightforward way that does not rely on seed area segmentation and avoid possible error accumulation brought by iterative processing, we propose a redundancy removal strategy. In this work, we directly explore two types of fuzzy semantics-text and separatrix-that do not possess specific boundaries, and separate cluttered instances by excluding the separatrix pixels from text regions. To deal with the fuzzy semantic boundaries, we also conduct reliability analysis in both optimization and inference stage to suppress false positive pixels at ambiguous locations. Experiments on benchmark datasets demonstrate the effectiveness of our method.

Intermediate feature maps of teacher model can produce enriched knowledge to improve the performance of student model. Existing works mainly focus on formulating beneficial knowledge for transferring, but ignore the contribution discrepancy of the knowledge to promote performance. To tackle this issue, we propose a simple Importance-based Knowledge Reweighting mechanism, which dynamically measure the importance of knowledge spatially and channel-wisely for teacher-student pairs. This reweighting scheme enables the student model to focus more on the prime knowledge. Furthermore, a local pattern consistency loss based on Structural Similarity Index Measure (SSIM) is presented to narrow the local pattern discrepancy between teacher and student features. Extensive experiments on CIFAR-100 with various combinations of network architectures for teacher and student well demonstrate the effectiveness and superiority of our proposed approach.

Knowledge distillation is an effective model compression technology, which encourages a small student model to mimic the features or probabilistic outputs of a large teacher model. Existing feature-based distillation methods mainly focus on formulating enriched representations, while naively address the channel dimension gap and adopt the handcrafted channel association strategy between teacher and student for distillation. This not only introduces more parameters and computational cost, but may transfer irrelevant information to student. In this paper, we present a differentiable and efficient Dynamic Channel Association (DCA) mechanism, which automatically associates proper teacher channels for each student channel. DCA also enables each student channel to distill knowledge from multiple teacher channels in a weighted manner. Extensive experiments on classification task, with various combinations of network architectures for teacher and student models, well demonstrate the effectiveness of our proposed approach.

Visual chirality measures the distribution variation of visual data under transformation, while it has not been explored in freehand sketches yet. In this paper, we investigate the vertical flipping associated with visual chirality in freehand sketches. Our analysis of investigation results reveals that the vertical flipping shows a high degree of visual chirality. To utilize the high-level cues automatically discovered by predicting the vertical flipping, we propose a Visual Chirality Attention (VCA) module for deep CNNs, which consists of two sequential sub-modules: channel and chirality attention. Experimental results of sketch recognition on TU-Berlin dataset show that our method performs more favorably against state-of-the-art attention-based methods. Our code can be found at https://github.com/zhengyinghit/VCANet.

Image decomposition is an important issue in image processing. The existing approaches including the bi‐dimensional empirical mode decomposition (BEMD) still fail to separate monocomponents in multicomponents in many cases. To solve this problem, this study proposes a new image decomposition method based on the new derived combined bi‐dimensional Bedrosian's principle that has not been reported anywhere else for image processing. First, this study investigates a few bi‐dimensional Bedrosian's principles according to the bi‐dimensional Hilbert transforms. Second, based on the derived bi‐dimensional Bedrosian's principles and the original multicomponents, the authors provide the combined bi‐dimensional Bedrosian's principle and the assisted components obtained through projections via optimisation so that these monocomponents in multicomponents can be separated in the case that the existing methods fail. Third, an iterative image decomposition method is proposed via the above principles to decompose the multicomponent image into true monocomponents. The proposed method can solve the problems caused by the cross‐angle and amplitude ratio and frequency ratio between these components that BEMD fails to solve. Also, the phase and amplitude are estimated for texture analysis after the decomposition is demonstrated. Experiments are shown to support the proposed methods.

In this study, some new entropic inequalities on sparse representation for pairs of bases are investigated. First, the generalised Shannon entropic uncertainty principle and the generalised Rényi entropic uncertainty principle via new derived Hausdorff–Young inequality are proved. These new derived uncertainty principles show that signals cannot have unlimited concentration related to minimum entropies in pairs of bases. Second, the conditions of uniqueness of sparse representation under minimum entropies are given. This study also demonstrates that the entropic greedy‐like algorithms can achieve the ‘sparsest’ representation for minimum entropies approximately. Third, the relations between the minimum l 0 solution and minimum entropy are discussed as well. It shows that even if the sparsest representation of l 0 ‐norm is obtained, the entropy cannot always be the minimum and it is possible that the entropy is limited to a interval. These new derived inequalities will be primarily to contribute to a better understanding of sparse representation in the sense of limited entropic uncertainty bounds. Finally, the experiments are shown to verify the authors’ ideas.

In this study, the authors investigated some new inequalities on sparse representation for pairs of bases and frames, which would enrich the theory ensemble. First, for fixed pairs of bases, frames and the signal to be represented, we presented the bounds (which can be used in practice directly) of l 0 ‐norm of signal coefficients by the max/min cross‐inner‐products, as is of much importance to bases and frames selections in sparse representation. Also, the error bounds associated with the given min cross‐inner‐products are achieved. Moreover, the equivalence condition between l 0 solution and l 1 solution was achieved via relations of cross‐inner‐products, as can be applied directly for selection of optimal bases. Finally, the estimation of the parameters of cross‐inner‐products was shown.