Representation Of Information Research Articles

Neural sensitivity to frequency changes in song structure in a high-order auditory area reflects tutor song memory in adult songbirds.

Vocal learners, including humans and songbirds, acquire their complex vocalizations by accurately memorizing and imitating the vocal patterns of other individuals. In songbirds, the caudomedial nidopallium (NCM), considered the secondary auditory region, has been suggested to play a critical role in memorizing and recognizing the songs of tutors. However, the mechanisms by which NCM neurons encode the acoustic information of tutor song are not yet fully understood. Here, we investigate the neural representation of tutor song information in NCM neurons by examining their sensitivity to spectral changes in song structure, using electrophysiological recordings in anesthetized male zebra finches. We manipulated the acoustic structures of both tutor songs and unfamiliar conspecific songs by shifting the fundamental frequency (FF) of harmonic syllables by various frequency steps and recorded neural responses to those FF-shifted and original songs. Our results demonstrate that NCM neurons are highly sensitive to FF shifts in tutor song but much less in unfamiliar conspecific song, providing novel evidence for neural encoding of tutor song information in NCM neurons. Moreover, we find that the effects of FF shifts on neural responses depend on the direction of FF shifts. These findings suggest that NCM neurons encode detailed information of tutor song, which can serve as a tutor song template required for song learning.

Event Argument Extraction for Rainstorm Disasters Based on Social Media: A Case Study of the 2021 Heavy Rains in Henan

Rainstorm disasters have wide-ranging impacts on communities, but traditional information collection methods are often hampered by high labor costs and limited coverage. Social media platforms such as Weibo provide new opportunities for monitoring and analyzing disaster-related information in real-time. In this paper, we present ETEN_BERT_QA, a novel model for extracting event arguments from Weibo rainstorm disaster texts. The model incorporates the event text enhancement network (ETEN) to enhance the extraction process by improving the semantic representation of event information in combination with event trigger words. To support our approach, we constructed RainEE, a dataset dedicated to rainstorm disaster event extraction, and implemented a two-step process, as follows: (1) event detection, which identifies trigger words and classifies them into event types, and (2) event argument extraction, which identifies event arguments and classifies them into argument roles. Our ETEN_BERT_QA model combines ETEN with a BERT-based question-answering mechanism to further improve the understanding of the event text. Experimental evaluations on the RainEE and DuEE datasets show that ETEN_BERT_QA significantly outperforms the baseline model in terms of accuracy and the number of event argument extractions, validating its effectiveness in analyzing rainstorm disaster-related Weibo texts.

Optimizing site selection for construction and demolition waste resource treatment plants using a hesitant neutrosophic set: a case study in Xiamen, China

The site selection of construction and demolition waste resource treatment plants is a complicated multi-criteria decision-making (MCDM) issue of ‘not in my back yard’ (NIMBY) facilities, which is affected by spatial, economic, societal, environmental and other factors. Traditional site selection MCDM methods expose the shortcomings of evaluation information representation and spatial factor analysis. This study develops an optimized MCDM site selection method in a hesitant neutrosophic set (HNS) environment, capable of effectively handling uncertainty and fully incorporating spatial considerations. The innovation of this research lies in integrating HNS into MCDM, combined with geographic information system technology, revised DEMATEL-based analytic network process method, cosine similarity and Euclidean distance similarity measures. A case study in Xiamen, China, validates this method’s effectiveness, and sensitivity analysis and comparative studies demonstrate its superior stability, robustness and resolution. The main contribution of this study is the development of an MCDM optimization method for siting large-scale NIMBY facilities.

Disentangled similarity graph attention heterogeneous biological memory network for predicting disease-associated miRNAs

BackgroundThe association between MicroRNAs (miRNAs) and diseases is crucial in treating and exploring many diseases or cancers. Although wet-lab methods for predicting miRNA-disease associations (MDAs) are effective, they are often expensive and time-consuming. Significant advancements have been made using Graph Neural Network-based methods (GNN-MDAs) to address these challenges. However, these methods still face limitations, such as not considering nodes’ deep-level similarity associations and hierarchical learning patterns. Additionally, current models do not retain the memory of previously learned heterogeneous historical information about miRNAs or diseases, only focusing on parameter learning without the capability to remember heterogeneous associations.ResultsThis study introduces the K-means disentangled high-level biological similarity to utilize potential hierarchical relationships fully and proposes a Graph Attention Heterogeneous Biological Memory Network architecture (DiGAMN) with memory capabilities. Extensive experiments were conducted across four datasets, comparing the DiGAMN model and its disentangling method against ten state-of-the-art non-disentangled methods and six traditional GNNs. DiGAMN excelled, achieving AUC scores of 96.35%, 96.10%, 96.01%, and 95.89% on the Data1 to Data4 datasets, respectively, surpassing all other models. These results confirm the superior performance of DiGAMN and its disentangling method. Additionally, various ablation studies were conducted to validate the contributions of different modules within the framework, and’s encoding statuses and memory units of DiGAMN were visualized to explore the utility and functionality of its modules. Case studies confirmed the effectiveness of DiGAMN’s predictions, identifying several new disease-associated miRNAs.ConclusionsDiGAMN introduces the use of a disentangled biological similarity approach for the first time and successfully constructs a Disentangled Graph Attention Heterogeneous Biological Memory Network model. This network can learn disentangled representations of similarity information and effectively store the potential biological entanglement information of miRNAs and diseases. By integrating disentangled similarity information with a heterogeneous attention memory network, DiGAMN enhances the model’s ability to capture and utilize complex underlying biological data, significantly outperforming many existing models. The concepts used in this method also provide new perspectives for predicting miRNAs associated with diseases.

Learning Disentangled Representation for One-Shot Progressive Face Swapping.

Although face swapping has attracted much attention in recent years, it remains a challenging problem. Existing methods leverage a large number of data samples to explore the intrinsic properties of face swapping without considering the semantic information of face images. Moreover, the representation of the identity information tends to be fixed, leading to suboptimal face swapping. In this paper, we present a simple yet efficient method named FaceSwapper, for one-shot face swapping based on Generative Adversarial Networks. Our method consists of a disentangled representation module and a semantic-guided fusion module. The disentangled representation module comprises an attribute encoder and an identity encoder, which aims to achieve the disentanglement of the identity and attribute information. The identity encoder is more flexible, and the attribute encoder contains more attribute details than its competitors. Benefiting from the disentangled representation, FaceSwapper can swap face images progressively. In addition, semantic information is introduced into the semantic-guided fusion module to control the swapped region and model the pose and expression more accurately. Experimental results show that our method achieves state-of-the-art results on benchmark datasets with fewer training samples.

Conti-Fuse: A novel continuous decomposition-based fusion framework for infrared and visible images

For better explore the relations of inter-modal and inner-modal, even in deep learning fusion framework, the concept of decomposition plays a crucial role. However, the previous decomposition strategies (base & detail or low-frequency & high-frequency) are too rough to present the common features and the unique features of source modalities, which leads to a decline in the quality of the fused images. The existing strategies treat these relations as a binary system, which may not be suitable for the complex generation task (e.g. image fusion). To address this issue, a continuous decomposition-based fusion framework (Conti-Fuse) is proposed. Conti-Fuse treats the decomposition results as few samples along the feature variation trajectory of the source images, extending this concept to a more general state to achieve continuous decomposition. This novel continuous decomposition strategy enhances the representation of complementary information of inter-modal by increasing the number of decomposition samples, thus reducing the loss of critical information. To facilitate this process, the continuous decomposition module (CDM) is introduced to decompose the input into a series continuous components. The core module of CDM, State Transformer (ST), is utilized to efficiently capture the complementary information from source modalities. Furthermore, a novel decomposition loss function is also designed which ensures the smooth progression of the decomposition process while maintaining linear growth in time complexity with respect to the number of decomposition samples. Extensive experiments demonstrate that our proposed Conti-Fuse achieves superior performance compared to the state-of-the-art fusion methods.

A multispatial information representation model emphasizing key brain regions for Alzheimer's disease diagnosis with structural magnetic resonance imaging.

Structural magnetic resonance imaging (sMRI) can reflect structural abnormalities of the brain. Due to its high tissue contrast and spatial resolution, it is considered as an MRI sequence in diagnostic tasks related to Alzheimer's disease (AD). Thus far, most studies based on sMRI have only focused on pathological changes in disease-related brain regions in Euclidean space, ignoring the association and interaction between brain regions represented in non-Euclidean space. This non-Euclidean spatial information can provide valuable information for brain disease research. However, few studies have combined Euclidean spatial information in images and graph spatial information in brain networks for the early diagnosis of AD. The purpose of this study is to explore how to effectively combine multispatial information for enhancing AD diagnostic performance. A multispatial information representation model (MSRNet) was constructed for the diagnosis of AD using sMRI. Specifically, the MSRNet included a Euclidean representation channel integrating a multiscale module and a feature enhancement module, in addition to a graph (non-Euclidean) representation channel integrating a node feature aggregation mechanism. This was accomplished through the adoption of a multilayer graph convolutional neural network and a node connectivity aggregation mechanism with fully connected layers. Each participants' gray-matter volume map and preconstructed radiomics-based morphology brain network (radMBN) were used as MSRNet inputs for the learning of multispatial information. Other than the multispatial information representation in MSRNet, an interactive mechanism was proposed to connect the Euclidean and graph representation channels by five disease-related brain regions which were identified based on a classifier operated on with two feature strategies of voxel intensities and radiomics features. MSRNet focused on disease-related brain regions while integrating multispatial information to effectively enhance disease discrimination. The MSRNet was validated on four publicly available datasets, achieving accuracies 92.8% and 90.6% for AD in intra-database and inter-database cross-validation, respectively. The accuracy of MSRNet in distinguishing between late mild cognitive impairment (MCI) and early MCI, and between progressive MCI and stable MCI, reached 79.8% and 73.4%, respectively. The experiments demonstrated that the model's decision scores exhibited good detection capability for MCI progression. Furthermore, the potential of decision scores for improving diagnostic performance was exhibited by combining decision scores with other clinical indicators for AD identification. The MSRNet model could conduct an effective multispatial information representation in the sMRI-based diagnosis of AD. The proposed interaction mechanism in the MSRNet could help the model focus on AD-related brain regions, thus further improving the diagnostic ability.

Representation of spatial information in the CA1 field

Information in the brain is encoded by large populations of neurons – neural ensembles. The place cells in the hippocampal CA1 field have become an experimental model for the study of neural ensembles of the brain due to the convenience of research. This review is devoted to the latest studies of place cells in the CA1 field. We consider the principles of encoding space by place cells, mechanisms for controlling the activity of place cells, anatomical and physiological features of place cells in different parts of the CA1 field. Key points: 1. There are rate and phase coding; 2. Dense local connections between pyramidal neurons can provide information processing; 3. Interneurons are involved in the formation of both the rate and phase code of place cells; 4. Pyramidal neurons are anatomically and functionally divided into deep and superficial; 5. Along the dorsoventral axis, the spatial and non-spatial component of information is generalized. The CA1 field has extensive capabilities for signal processing and can implement a computationally complex operation in the cognitive processes of the brain.

Semantic decomposition and enhancement hashing for deep cross-modal retrieval

Deep hashing has garnered considerable interest and has shown impressive performance in the domain of retrieval. However, the majority of the current hashing techniques rely solely on binary similarity evaluation criteria to assess the semantic relationships between multi-label instances, which presents a challenge in overcoming the feature gap across various modalities. In this paper, we propose semantic decomposition and enhancement hashing (SDEH) by extensively exploring the multi-label semantic information shared by different modalities for cross-modal retrieval. Specifically, we first introduce two independent attention-based feature learning subnetworks to capture the modality-specific features with both global and local details. Subsequently, we exploit the semantic features from multi-label vectors by decomposing the shared semantic information among multi-modal features such that the associations of different modalities can be established. Finally, we jointly learn the common hash code representations of multimodal information under the guidelines of quadruple losses, making the hash codes informative while simultaneously preserving multilevel semantic relationships and feature distribution consistency. Comprehensive experiments on four commonly used multimodal datasets offer strong support for the exceptional effectiveness of our proposed SDEH.

GDTE-based crack diagnosis for planetary gear: Mechanism, validation, and advantages compared to vibration-based technology

Effective health monitoring and fault diagnosis technologies are crucial to timely grasping the operation status of the planetary gear train (PGT). In this study, the mechanism for global dynamic transmission error (GDTE)-based crack diagnosis of PGT is revealed from dynamic responses and validated through experiments, and the advantages of this new technique are compared with the commonly used vibration-based technology. Initially, a dynamic model of the PGT considering various crack degrees and transfer path effects is established, and the effectiveness of the model response is verified through experiment. Subsequently, GDTE and vibration signals under different crack conditions are statistically analyzed in time domain and frequency domain to evaluate the differences in the ability to reveal gear failure information. Ultimately, the reasons for the different distribution of fault characterization information in the two signals are explained from a dynamic perspective. The results show that GDTE-based fault diagnosis exhibits greater sensitivity to gear crack degree compared to vibration-based monitoring technology, with both time and frequency domain responses containing richer fault characteristic information, particularly in low-frequency regions. The fault characteristics concentrated in higher-frequency regions in the vibration signal are mainly caused by the modulation between gear crack shock and the gear single and double tooth alternating meshing shock. Conversely, fault frequencies in GDTE signals are mainly found in low-frequency regions, as fault shocks in GDTE signals directly correlate with the rotational frequency of the cracked gear. Modulation effects and attenuation on the transfer path are identified as the main reasons for the weaker representation of gear fault information in vibration signals compared to GDTE signals.

Action similarity warps visual feature space in working memory

Visual working memory (VWM) retains representations of past visual information for future action. Yet to date, most studies have approached VWM as just serving perception beyond the immediate. Whether and how prospective actions shape information in VWM remains largely unknown, in part because typical experimental setups limit behavior to simple button presses. In two experiments (one preregistered), using a novel interactive VWM task, we show that the similarity of the actions that we intend to perform on memory items adaptively distorts their representation. Participants memorized the orientation of two bars, after which they were informed as to which manual actions they should reproduce these orientations with in a memory recall test. We observed that perceptually similar items were remembered as more distinct when paired with different action plans versus the same action plan. A control experiment showed that this action-induced effect reflects a true change in the visual representation rather than a motor bias. These findings demonstrate that VWM representations are flexibly adapted to guide specific overt actions and provide evidence that action plans can retrospectively warp sensory feature space in VWM.

A temporally insensitive spatio-temporal fusion method for remote sensing imagery via semantic prior regularization

Spatio-temporal fusion has become a popular technology for generating remote sensing images with high spatial and high temporal resolutions, thus providing valuable data support for remote sensing monitoring applications, such as environmental monitoring and city planning. Currently, deep learning-based methods have garnered a significant amount of attention, and they mostly employ the fine image at the neighboring date as an auxiliary image. However, capturing usable neighboring fine images may be challenging due to the adverse effects of weather conditions on optical images. Moreover, the fusion performance drops sharply when the temporal interval is long (i.e., there are significant differences in images). In this paper, we proposed a bidirectional pyramid fusion network with semantic prior regularization (BPFN-SPR), which exhibits remarkable flexibility and robustness to temporal intervals.Specifically, the proposed BPFN-SPR contains dual-path operations (i.e., Semantic Extraction path and Image Reconstruction path). The semantic extraction path has two modes: parameter learning mode and parameter freezing mode. The parameter learning mode aims to learn the information representation of the auxiliary fine image, while the parameter freezing mode aims to perceive the accurate semantic information of the target fine image. The image reconstruction path progressively reconstructs spatial details of fine images from coarse images, which jointly optimizes the target fine image and the auxiliary fine image, reducing the temporal sensitivity of the reconstruction branch, and thereby improving its generalization ability. Experimental results show that the proposed method has competitive performance, especially for areas with land cover changes. In addition, extensive experiments using images at multi-temporal intervals as auxiliary images have also demonstrated the significant advantages of the proposed method. The mean PSNR value attains 31.0713, while the average spectral index SAM measures 0.1640 on the LGC test set. Meanwhile, for the CIA test set, the average PSNR is recorded at 29.5332, accompanied by an average spectral index SAM of 0.1865. Therefore, the proposed BPFN-SPR has considerable potential in monitoring Earth's surface dynamics.

“Gender matters”: the development of infographics to raise awareness and promote gender-transformative care in traumatic brain injury

ABSTRACT Purpose To develop a series of infographics providing persons with traumatic brain injury (TBI) and their circle of care with evidence-based information on sex and gender topics in TBI. Materials and Methods We employed an iterative participatory design engaging knowledge users, scientists, and experts in brain injury and patient education. To inform infographic content, we conducted an information needs assessment with knowledge users through semi-structured interviews and referred to our previously published evidence syntheses on TBI topics. We followed principles of graphic design and science communication to create materials reflecting lived experiences of knowledge users. Results We created a series of infographics with actionable messages and visual representations of evidence-based information. We achieved a Flesch Reading-Ease score of 60.1, corresponding to a Grade 7/8 reading level. The infographics met the color contrast criteria of the Web Content Accessibility Guidelines. Knowledge users found the material useful, visually appealing, and helpful in understanding complex topics. Conclusions There is value in merging art and science to develop educational materials that meet the unique information needs of knowledge users. Iterative participatory design engaging diverse stakeholders is essential for co-creating knowledge translation tools to improve access to health information and quality of care after TBI.

Representational learning by optimization of neural manifolds in an olfactory memory network.

Higher brain functions depend on experience-dependent representations of relevant information that may be organized by attractor dynamics or by geometrical modifications of continuous "neural manifolds". To explore these scenarios we analyzed odor-evoked activity in telencephalic area pDp of juvenile and adult zebrafish, the homolog of piriform cortex. No obvious signatures of attractor dynamics were detected. Rather, olfactory discrimination training selectively enhanced the separation of neural manifolds representing task-relevant odors from other representations, consistent with predictions of autoassociative network models endowed with precise synaptic balance. Analytical approaches using the framework of manifold capacity revealed multiple geometrical modifications of representational manifolds that supported the classification of task-relevant sensory information. Manifold capacity predicted odor discrimination across individuals, indicating a close link between manifold geometry and behavior. Hence, pDp and possibly related recurrent networks store information in the geometry of representational manifolds, resulting in joint sensory and semantic maps that may support distributed learning processes.

Differential neural representations of syntactic and semantic information across languages in Chinese-English bilinguals

Bilingual individuals manage multiple languages that align in conceptual meaning but differ in forms and structures. While prior research has established foundational insights into the neural mechanisms in bilingual processing, the extent to which the first (L1) and second language (L2) systems overlap or diverge across different linguistic components remains unclear. This study probed the neural underpinnings of syntactic and semantic processing for L1 and L2 in Chinese-English bilinguals (N = 44) who performed sentence comprehension tasks and an N-back working memory task during functional MRI scanning. We observed that the increased activation for L2 processing was within the verbal working memory network, suggesting a greater cognitive demand for processing L2. Crucially, we looked for brain regions showing adaptation to the repetition of semantic information and syntactic structure, and found more robust adaptation effects in L1 in the middle and superior temporal cortical areas. The differential adaptation effects between L1 and L2 were more pronounced for the semantic condition. Multivariate pattern analysis further revealed distinct neural sensitivities to syntactic and semantic representations between L1 and L2 across frontotemporal language regions. Our findings suggest that while L1 and L2 engage similar neural systems, finer representation analyses uncover distinct neural patterns for both semantic and syntactic aspects in the two languages. This study advances our understanding of neural representations involved in different language components in bilingual individuals.

A tactile perception method with flexible grating structural color

Abstract Affordable high-resolution cameras and state-of-the-art computer vision techniques have led to the emergence of various vision-based tactile sensors. However, current vision-based tactile sensors mainly depend on geometric optics or marker tracking for tactile assessments, resulting in limited performance. To solve this dilemma, we introduce optical interference patterns as the visual representation of tactile information for flexible tactile sensors. We propose a novel tactile perception method and its corresponding sensor combining structural colors from flexible blazed gratings with deep learning. The richer structural colors and finer data processing foster the tactile estimation performance. The proposed sensor has an overall normal force magnitude accuracy of 6 mN, a planar resolution of 79 μm, and a contact-depth resolution of 25 μm. This work presents a promising tactile method that combines wave optics, soft materials, and machine learning. It demonstrates well in tactile measurement, and can be expanded into multiple sensing fields.

FAQ-Gen: An automated system to generate domain-specific FAQs to aid content comprehension

Frequently Asked Questions (FAQs) refer to the most common inquiries about specific content. They serve as content comprehension aids by simplifying topics and enhancing understanding through succinct presentation of information. In this paper, we address FAQ generation as a well-defined Natural Language Processing task through the development of an end-to-end system leveraging text-to-text transformation models. We present a literature review covering traditional question-answering systems, highlighting their limitations when applied directly to the FAQ generation task. We propose a system capable of building FAQs from textual content tailored to specific domains, enhancing their accuracy and relevance. We utilise self-curated algorithms to obtain an optimal representation of information to be provided as input and also to rank the question-answer pairs to maximise human comprehension. Qualitative human evaluation showcases the generated FAQs as well-constructed and readable while also utilising domain-specific constructs to highlight domain-based nuances and jargon in the original content.

Image forgery localization integrating multi-scale and boundary features

Abstract Image forgery localization identifies tampered regions within an image by extracting distinctive forgery features. Current methods mainly use convolutional neural networks (CNNs) to extract features. However, CNNs’ limited receptive field emphasizes local features, impeding the global modeling of crucial lower-level features like edges and textures, leading to decreased precision. Moreover, prior methods use pyramid networks for multi-scale feature extraction but show deficiencies in multi-scale and interlayer modeling, leading to inadequate multi-scale information representation and limiting flexibility to tampered regions of varying sizes. To address these issues, this paper proposes a Transformer-based model integrating multi-scale and boundary features. The model employs a Pyramid Vision Transformer as the encoder, using self-attention over convolution to enhance global context modeling. Building on this, the model incorporates a multi-scale feature enhancement module that enriches forgery features by paralleling various convolutional layers. Features at various encoder stages are integrated through a cross-stage interaction module, enabling multi-level feature correlation for a strong feature representation. Furthermore, the model includes a forgery boundary information-guided branch, which focuses precisely on tampered region structures without introducing irrelevant noise. Experiments demonstrate that our model surpasses previous methods in localization accuracy, with F1 and AUC improving by 8.5% and 2.2% in pre-training, respectively.

Intelligent fault diagnosis in power distribution networks using LSTM-DenseNet network

This paper introduces a novel fault diagnosis method that combines DenseNet and Long Short-Term Memory (LSTM) networks. The DenseNet utilizes its unique dense block structure to detect subtle variations in three-phase voltage and zero-sequence current signals. In addition, the Squeeze-and-Excitation (SE) module is introduced in DenseNet. The SE module enhances DenseNet's feature representation by adapting the importance of each channel in the feature map. Furthermore, integrating the LSTM model enables capturing time-domain features of fault signals, enhancing the analysis of waveform changes and trends. These extracted features are subsequently fused in a cascaded manner, leveraging the strengths of both approaches to obtain a more comprehensive information representation. To better explain the capability of feature extraction in each part of the model, t-distributed Stochastic Neighbor Embedding (t-SNE) method is used for visual analysis. The proposed method is evaluated using two distribution network models, namely the 10 kV and IEEE34 networks, in simulation. The verification results indicate that the proposed method achieves exceptionally high accuracy in fault identification for both tested distribution network models, with rates of 99.87 % and 99.82 %, respectively, while also demonstrating robust performance in noisy environments. This performance surpasses that of other related methods, underscoring the enhanced effectiveness of our approach.

A Dual-Stream Fusion Network for Human Energy Expenditure Estimation with Wearable Sensor

With the increasing awareness of health, using wearable sensors to monitor individual activities and accurately estimate energy expenditure has become a current research focus. However, existing research encounters challenges including low estimation accuracy, a deficiency of frequency domain features, and difficulty in integrating time domain and frequency domain features. To address these issues, we propose an innovative framework called the Dual-Stream Fusion Network (DSFN). This framework combines the Time Domain Encoding (TDE) module, the Frequency Domain Hierarchical-Split Encoding (FDHSE) module, and a Two-Stage Feature Fusion (TSF) module. Specifically, the temporal stream of the framework employs the TDE module to capture deep temporal features that reflect the complex dynamic variations in time-series data. The frequency domain stream introduces the FDHSE module, which extracts frequency domain features using a multi-level, multi-scale approach, ensuring a comprehensive and diverse representation of frequency information. Through this dual-stream architecture, our model effectively learns both time and frequency domain features, addressing the limitations of frequency domain features observed in prior studies. Additionally, we propose the TSF module to fully integrate time and frequency domain features, effectively overcoming the challenge of fusing these two types of features. We conducted experiments on two public datasets, namely the GOTOV dataset (elderly people) and the JSI dataset (young people). Experimental results demonstrate that our method achieves excellent performance across different age groups. Compared to the baseline models, the proposed DSFN significantly improves the accuracy of human energy expenditure estimation.