Dynamic sign language recognition holds significant importance, particularly with the application of deep learning to address its complexity. However, existing methods face several challenges. Firstly, recognizing dyn...Dynamic sign language recognition holds significant importance, particularly with the application of deep learning to address its complexity. However, existing methods face several challenges. Firstly, recognizing dynamic sign language requires identifying keyframes that best represent the signs, and missing these keyframes reduces accuracy. Secondly, some methods do not focus enough on hand regions, which are small within the overall frame, leading to information loss. To address these challenges, we propose a novel Video Transformer Attention-based Network (VTAN) for dynamic sign language recognition. Our approach prioritizes informative frames and hand regions effectively. To tackle the first issue, we designed a keyframe extraction module enhanced by a convolutional autoencoder, which focuses on selecting information-rich frames and eliminating redundant ones from the video sequences. For the second issue, we developed a soft attention-based transformer module that emphasizes extracting features from hand regions, ensuring that the network pays more attention to hand information within sequences. This dual-focus approach improves effective dynamic sign language recognition by addressing the key challenges of identifying critical frames and emphasizing hand regions. Experimental results on two public benchmark datasets demonstrate the effectiveness of our network, outperforming most of the typical methods in sign language recognition tasks.展开更多
Azalea is a general designation of Rhododendron in the Ericaceae family.Rhododendron not only has high ornamental value but also has application value in ecological protection,medicine,and scientific research.In this ...Azalea is a general designation of Rhododendron in the Ericaceae family.Rhododendron not only has high ornamental value but also has application value in ecological protection,medicine,and scientific research.In this study,we used Illumina and PacBio sequencing to assemble and annotate the entire chloroplast genomes(cp genomes)of four Rhododendron species.The chloroplast genomes of R.concinnum,R.henanense subsp.lingbaoense,R.micranthum,and R.simsii were assembled into 207,236,208,015,207,233,and 206,912 bp,respectively.All chloroplast genomes contain eight rRNA genes,with either 88 or 89 protein-coding genes.The four cp genomes were compared and analyzed by bioinformatics,and the phylogenetic analysis based on chloroplast genomes of 26 species of Ericaceae,Actinidiaceae,and Primulaceae under Ericales was conducted.A comparison of the linear structure of cp genomes of four Rhododendron showed that there were substantial sequence similarities in coding regions,but high differences in non-coding regions.A phylogenetic analysis,based on chloroplast whole genome sequences,showed that all Rhododendron species are in the clade Ericaceae.This study provides valuable genetic information for the study of population genetics and evolutionary relationships in Rhododendron and other azalea species.展开更多
Smell that exists in the natural environment is composed of numerous odor molecules(Bushdid et al.,2014).The mammalian olfactory system can accurately identify environmental olfactory cues,including those related to f...Smell that exists in the natural environment is composed of numerous odor molecules(Bushdid et al.,2014).The mammalian olfactory system can accurately identify environmental olfactory cues,including those related to food selection,recognition of conspecifics/predators,and emotional responses.展开更多
Brain signals refer to electrical signals or metabolic changes that occur as a consequence of brain cell activity.Among the various non-invasive measurement methods,electroencephalogram(EEG)stands out as a widely empl...Brain signals refer to electrical signals or metabolic changes that occur as a consequence of brain cell activity.Among the various non-invasive measurement methods,electroencephalogram(EEG)stands out as a widely employed technique,providing valuable insights into brain patterns.The deviations observed in EEG reading serve as indicators of abnormal brain activity,which is associated with neurological diseases.Brain‒computer interface(BCI)systems enable the direct extraction and transmission of information from the human brain,facilitating interaction with external devices.Notably,the emergence of artificial intelligence(AI)has had a profound impact on the enhancement of precision and accuracy in BCI technology,thereby broadening the scope of research in this field.AI techniques,encompassing machine learning(ML)and deep learning(DL)models,have demonstrated remarkable success in classifying and predicting various brain diseases.This comprehensive review investigates the application of AI in EEG-based brain disease diagnosis,highlighting advancements in AI algorithms.展开更多
Bitter receptors were initially identified within the gustatory system.In recent years,bitter receptors have been found in various non-gustatory tissues,including the cardiovascular system,where they participate in di...Bitter receptors were initially identified within the gustatory system.In recent years,bitter receptors have been found in various non-gustatory tissues,including the cardiovascular system,where they participate in diverse physiological processes.To investigate the electrophysiological and potential therapeutic implications of bitter receptors,we have developed a highly sensitive,multifunctional planar-electroporated cell biosensor(PECB)for high-throughput evaluation of the effects of bitter substances on cardiomyocytes.The PECB demonstrated the capability for highthroughput,stable,and reproducible detection of intracellular action potentials(IAPs).In comparison to conventional biosensors that utilize extracellular action potentials(EAPs)for data analysis,the IAPs recorded by the PECB provided high-resolution insights into action potentials,characterized by increased amplitudes and an enhanced signal-to-noise ratio(SNR).The PECB successfully monitored IAPs induced by the activation of bitter receptors by using three bitter substances:diphenidol,denatonium benzoate,and arbutin in cardiomyocytes.To further assess the drug development ability of our PECB,we established an in vitro long QT syndrome(LQTS)model to investigate the potential therapeutic effects of arbutin.The results indicated that arbutin altered the electrophysiological properties of cardiomyocytes and significantly shortened the repolarization time in the LQTS model.Moreover,it demonstrated its potential mechanistic pathway by activating bitter receptors to modulate cardiac ion channel activities.The developed PECB provides an effective platform for high-throughput screening of substrates of bitter receptors for the treatment of heart disease,presenting new opportunities for the development of antiarrhythmic therapies.展开更多
Full waveform inversion(FWI) is a seismic imaging method with a unified mathematical framework based on wave equation constraints. The FWI method can be used to generate a variety of high-resolution seismic parameter ...Full waveform inversion(FWI) is a seismic imaging method with a unified mathematical framework based on wave equation constraints. The FWI method can be used to generate a variety of high-resolution seismic parameter models(e.g.,velocity, anisotropy, viscoelasticity, and attenuation), which can facilitate an in-depth understanding of important scientific problems such as the Earth's interior structure and material composition, earthquake preparation and occurrence, and plate motion and dynamic processes. With the development and cross-integration of disciplines such as geophysics, applied mathematics, and computer science, FWI imaging theories and methods not only play a crucial role in revealing the Earth's interior structure, dynamic evolution, and earthquake mechanisms but also show a wide range of application potential in fields such as resource exploration, medical imaging, engineering inspection, carbon dioxide geological sequestration, and earthquake disaster prediction. In this paper, we provide a comprehensive review and analysis of the development of the FWI method,addressing its current challenges, identifying key issues, future directions, and potential research areas in the theory, methodology, and application of high-resolution FWI imaging. We also offer new insights and perspectives to promote advancements of high-resolution FWI research and applications in Earth sciences and other related fields.展开更多
基金supported by the National Natural Science Foundation of China under Grant Nos.62076117 and 62166026the Jiangxi Provincial Key Laboratory of Virtual Reality under Grant No.2024SSY03151.
文摘Dynamic sign language recognition holds significant importance, particularly with the application of deep learning to address its complexity. However, existing methods face several challenges. Firstly, recognizing dynamic sign language requires identifying keyframes that best represent the signs, and missing these keyframes reduces accuracy. Secondly, some methods do not focus enough on hand regions, which are small within the overall frame, leading to information loss. To address these challenges, we propose a novel Video Transformer Attention-based Network (VTAN) for dynamic sign language recognition. Our approach prioritizes informative frames and hand regions effectively. To tackle the first issue, we designed a keyframe extraction module enhanced by a convolutional autoencoder, which focuses on selecting information-rich frames and eliminating redundant ones from the video sequences. For the second issue, we developed a soft attention-based transformer module that emphasizes extracting features from hand regions, ensuring that the network pays more attention to hand information within sequences. This dual-focus approach improves effective dynamic sign language recognition by addressing the key challenges of identifying critical frames and emphasizing hand regions. Experimental results on two public benchmark datasets demonstrate the effectiveness of our network, outperforming most of the typical methods in sign language recognition tasks.
基金supported by the National Natural Science Foundation of China[Grant No.31870697].
文摘Azalea is a general designation of Rhododendron in the Ericaceae family.Rhododendron not only has high ornamental value but also has application value in ecological protection,medicine,and scientific research.In this study,we used Illumina and PacBio sequencing to assemble and annotate the entire chloroplast genomes(cp genomes)of four Rhododendron species.The chloroplast genomes of R.concinnum,R.henanense subsp.lingbaoense,R.micranthum,and R.simsii were assembled into 207,236,208,015,207,233,and 206,912 bp,respectively.All chloroplast genomes contain eight rRNA genes,with either 88 or 89 protein-coding genes.The four cp genomes were compared and analyzed by bioinformatics,and the phylogenetic analysis based on chloroplast genomes of 26 species of Ericaceae,Actinidiaceae,and Primulaceae under Ericales was conducted.A comparison of the linear structure of cp genomes of four Rhododendron showed that there were substantial sequence similarities in coding regions,but high differences in non-coding regions.A phylogenetic analysis,based on chloroplast whole genome sequences,showed that all Rhododendron species are in the clade Ericaceae.This study provides valuable genetic information for the study of population genetics and evolutionary relationships in Rhododendron and other azalea species.
基金supported by the Zhejiang Provincial Natural Science Foundation of China (Nos.LY21C100001 and LBY21H180001)the National Natural Science Foundation of China (Nos.62271443 and 32250008).
文摘Smell that exists in the natural environment is composed of numerous odor molecules(Bushdid et al.,2014).The mammalian olfactory system can accurately identify environmental olfactory cues,including those related to food selection,recognition of conspecifics/predators,and emotional responses.
基金supported by the National Key Research and Development Project of China(No.2021ZD0200405)the National Natural Science Foundation of China(Nos.62271443,32250008,and 82330064).
文摘Brain signals refer to electrical signals or metabolic changes that occur as a consequence of brain cell activity.Among the various non-invasive measurement methods,electroencephalogram(EEG)stands out as a widely employed technique,providing valuable insights into brain patterns.The deviations observed in EEG reading serve as indicators of abnormal brain activity,which is associated with neurological diseases.Brain‒computer interface(BCI)systems enable the direct extraction and transmission of information from the human brain,facilitating interaction with external devices.Notably,the emergence of artificial intelligence(AI)has had a profound impact on the enhancement of precision and accuracy in BCI technology,thereby broadening the scope of research in this field.AI techniques,encompassing machine learning(ML)and deep learning(DL)models,have demonstrated remarkable success in classifying and predicting various brain diseases.This comprehensive review investigates the application of AI in EEG-based brain disease diagnosis,highlighting advancements in AI algorithms.
基金supported by the National Key Research and Development Program of China(2024YFB3212300)Key Project of Zhejiang Province(2023C03104,2024C03146)+4 种基金National Natural Science Foundation of China(32201082,62301481,62401505)Scientific Research Fund of Zhejiang Provincial Education Department(Y202353232)the Postdoctoral Fellowship Program of CPSF(GZC20232333)Postdoctoral Science Foundation Funded Project(BX2021265,2021M702859)the Fundamental Research Funds for the Central Universities(226-2024-00059).
文摘Bitter receptors were initially identified within the gustatory system.In recent years,bitter receptors have been found in various non-gustatory tissues,including the cardiovascular system,where they participate in diverse physiological processes.To investigate the electrophysiological and potential therapeutic implications of bitter receptors,we have developed a highly sensitive,multifunctional planar-electroporated cell biosensor(PECB)for high-throughput evaluation of the effects of bitter substances on cardiomyocytes.The PECB demonstrated the capability for highthroughput,stable,and reproducible detection of intracellular action potentials(IAPs).In comparison to conventional biosensors that utilize extracellular action potentials(EAPs)for data analysis,the IAPs recorded by the PECB provided high-resolution insights into action potentials,characterized by increased amplitudes and an enhanced signal-to-noise ratio(SNR).The PECB successfully monitored IAPs induced by the activation of bitter receptors by using three bitter substances:diphenidol,denatonium benzoate,and arbutin in cardiomyocytes.To further assess the drug development ability of our PECB,we established an in vitro long QT syndrome(LQTS)model to investigate the potential therapeutic effects of arbutin.The results indicated that arbutin altered the electrophysiological properties of cardiomyocytes and significantly shortened the repolarization time in the LQTS model.Moreover,it demonstrated its potential mechanistic pathway by activating bitter receptors to modulate cardiac ion channel activities.The developed PECB provides an effective platform for high-throughput screening of substrates of bitter receptors for the treatment of heart disease,presenting new opportunities for the development of antiarrhythmic therapies.
基金supported by the Key Project of the National Natural Science Foundation of China (Grant No. 42330801)。
文摘Full waveform inversion(FWI) is a seismic imaging method with a unified mathematical framework based on wave equation constraints. The FWI method can be used to generate a variety of high-resolution seismic parameter models(e.g.,velocity, anisotropy, viscoelasticity, and attenuation), which can facilitate an in-depth understanding of important scientific problems such as the Earth's interior structure and material composition, earthquake preparation and occurrence, and plate motion and dynamic processes. With the development and cross-integration of disciplines such as geophysics, applied mathematics, and computer science, FWI imaging theories and methods not only play a crucial role in revealing the Earth's interior structure, dynamic evolution, and earthquake mechanisms but also show a wide range of application potential in fields such as resource exploration, medical imaging, engineering inspection, carbon dioxide geological sequestration, and earthquake disaster prediction. In this paper, we provide a comprehensive review and analysis of the development of the FWI method,addressing its current challenges, identifying key issues, future directions, and potential research areas in the theory, methodology, and application of high-resolution FWI imaging. We also offer new insights and perspectives to promote advancements of high-resolution FWI research and applications in Earth sciences and other related fields.
