Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
Objective Repetitive transcranial magnetic stimulation(rTMS)has demonstrated efficacy in enhancing neurocognitive performance in Alzheimer’s disease(AD),but the neurobiological mechanisms linking synaptic pathology,n...Objective Repetitive transcranial magnetic stimulation(rTMS)has demonstrated efficacy in enhancing neurocognitive performance in Alzheimer’s disease(AD),but the neurobiological mechanisms linking synaptic pathology,neural oscillatory dynamics,and brain network reorganization remain unclear.This investigation seeks to systematically evaluate the therapeutic potential of rTMS as a non-invasive neuromodulatory intervention through a multimodal framework integrating clinical assessments,molecular profiling,and neurophysiological monitoring.Methods In this prospective double-blind trial,12 AD patients underwent a 14-day protocol of 20 Hz rTMS,with comprehensive multimodal assessments performed pre-and postintervention.Cognitive functioning was quantified using the mini-mental state examination(MMSE)and Montreal cognitive assessment(MOCA),while daily living capacities and neuropsychiatric profiles were respectively evaluated through the activities of daily living(ADL)scale and combined neuropsychiatric inventory(NPI)-Hamilton depression rating scale(HAMD).Peripheral blood biomarkers,specifically Aβ1-40 and phosphorylated tau(p-tau181),were analyzed to investigate the effects of rTMS on molecular metabolism.Spectral power analysis was employed to investigate rTMS-induced modulations of neural rhythms in AD patients,while brain network analyses incorporating topological properties were conducted to examine stimulus-driven network reorganization.Furthermore,systematic assessment of correlations between cognitive scale scores,blood biomarkers,and network characteristics was performed to elucidate cross-modal therapeutic associations.Results Clinically,MMSE and MOCA scores improved significantly(P<0.05).Biomarker showed that Aβ1-40 level increased(P<0.05),contrasting with p-tau181 reduction.Moreover,the levels of Aβ1-40 were positively correlated with MMSE and MOCA scores.Post-intervention analyses revealed significant modulations in oscillatory power,characterized by pronounced reductions in delta(P<0.05)and theta bands(P<0.05),while concurrent enhancements were observed in alpha,beta,and gamma band activities(all P<0.05).Network analysis revealed frequency-specific reorganization:clustering coefficients were significantly decreased in delta,theta,and alpha bands(P<0.05),while global efficiency improvement was exclusively detected in the delta band(P<0.05).The alpha band demonstrated concurrent increases in average nodal degree(P<0.05)and characteristic path length reduction(P<0.05).Further research findings indicate that the changes in the clinical scale HAMD scores before and after rTMS stimulation are negatively correlated with the changes in the blood biomarkers Aβ1-40 and p-tau181.Additionally,the changes in the clinical scales MMSE and MoCA scores were negatively correlated with the changes in the node degree of the alpha frequency band and negatively correlated with the clustering coefficient of the delta frequency band.However,the changes in MMSE scores are positively correlated with the changes in global efficiency of both the delta and alpha frequency bands.Conclusion 20 Hz rTMS targeting dorsolateral prefrontal cortex(DLPFC)significantly improves cognitive function and enhances the metabolic clearance ofβ-amyloid and tau proteins in AD patients.This neurotherapeutic effect is mechanistically associated with rTMS-mediated frequency-selective neuromodulation,which enhances the connectivity of oscillatory networks through improved neuronal synchronization and optimized topological organization of functional brain networks.These findings not only support the efficacy of rTMS as an adjunctive therapy for AD but also underscore the importance of employing multiple assessment methods—including clinical scales,blood biomarkers,and EEG——in understanding and monitoring the progression of AD.This research provides a significant theoretical foundation and empirical evidence for further exploration of rTMS applications in AD treatment.展开更多
Background:Parkinson’s disease(PD)is one of the most common movement disorders worldwide.Ziyin Xifeng Decoction(ZYXFD),a traditional Chinese medicine compound formula,has shown therapeutic efficacy in treating PD,but...Background:Parkinson’s disease(PD)is one of the most common movement disorders worldwide.Ziyin Xifeng Decoction(ZYXFD),a traditional Chinese medicine compound formula,has shown therapeutic efficacy in treating PD,but its specific mechanisms of action have not been fully elucidated.Methods:Firstly,we employed network pharmacology and untargeted metabolomics analysis to identify the core targets,pathways,and key metabolites of ZYXFD in the treatment of PD.Subsequently,we evaluated the protective effects of ZYXFD and further investigated its anti-PD mechanisms by validating the analytical results.Results:Combined analyses of network pharmacology and metabolomics identify the core targets including EGFR,SRC,PTGS2,and CDK2,while the effects of ZYXFD against PD are likely mediated primarily through the PI3K/AKT/mTOR signaling pathway.Pharmacodynamic evaluation demonstrated that a high dose of ZYXFD significantly improved behavioral deficits in chronic PD mice,downregulatedα-synuclein protein expression,and protected dopaminergic neurons.It also regulated the expression of core targets,inhibited the PI3K/AKT/mTOR signaling pathway,promoted autophagy,and reduced apoptosis.In vitro experiments further verified that the therapeutic effect of ZYXFD on PD is dependent on autophagy regulation.Conclusion:The findings demonstrated that ZYXFD alleviates PD by modulating related proteins and metabolites,inhibiting the PI3K/AKT/mTOR signaling pathway,and enhancing autophagy.This provides a theoretical basis for its broader application in PD treatment.展开更多
Mitochondrial DNA variants have been linked to cognitive progression in Parkinson’s disease;however,the mechanisms by which mitochondrial DNA variants or haplogroups contribute to this process remain unclear.In the p...Mitochondrial DNA variants have been linked to cognitive progression in Parkinson’s disease;however,the mechanisms by which mitochondrial DNA variants or haplogroups contribute to this process remain unclear.In the present study,we analyzed single-nucleus RNA sequencing data from 241 post-mortem brain samples across five regions to investigate the dysregulatory mechanisms associated with mitochondrial DNA haplogroup H and haplogroups J,T,and U#.Our findings revealed significant alterations in the proportions of astrocyte subtypes CHI3L1 and GRM3 in the neocortical regions of haplogroup H.Notably,TTR was markedly downregulated in the dorsal motor nucleus of the Xth nerve region of patients with haplogroup H.Pathway analysis highlighted abnormal hypoxic and reactive oxygen species environments in astrocytes,whereas protein complex analysis revealed a consistent and significant elevation in ribosomal subunit complexes within the astrocyte subtypes.By constructing weighted and directed transcriptome-wide gene regulatory networks,we identified significant changes in transcription factor SP1 and homeobox protein HOXA5 activity in the astrocyte subtypes of individuals with haplogroup H.Additionally,widespread dysregulation was observed in the transcriptional control of TTR by multiple transcription factors.Parkinson’s disease patients with haplogroup H also exhibited increased network functional connectivity in specific brain regions.This data-driven study underscores the potential mechanisms by which mitochondrial DNA haplogroups contribute to cognitive progression in Parkinson’s disease,involving cellular composition changes,differential gene expression,pathway disruption,and gene regulatory networks.Our findings suggest that mitochondrial DNA haplogroup H may drive Parkinson’s disease cognitive progression through aberrant TTR expression and a hypoxic environment.展开更多
The 200 Gbit/s passive optical network(PON)is most likely to be the next-generation scheme following 50G PON.The costeffective direct detection(DD)system is the economical choice.However,larger-capacity DD systems wil...The 200 Gbit/s passive optical network(PON)is most likely to be the next-generation scheme following 50G PON.The costeffective direct detection(DD)system is the economical choice.However,larger-capacity DD systems will face much more serious power fading caused by chromatic dispersion(CD)combined with square-law DD and thereby significantly increases the complexity of equalization algorithms.In this paper,a 200 Gbit/s Nyquist 4-level pulse amplitude modulation(PAM4)single side-band(SSB)modulation-DD downlink scheme is designed,and a low complexity quadratic-nonlinear equalizer is proposed for this system.The computational complexity of the quadratic nonlinear equalizer is about 28%of that of the conventional Volterra nonlinear equalizer,while still exhibiting excellent nonlinear equalization ability.Simulation results for the 200 Gbit/s system with 20 km fiber transmission show that it can achieve a power budget of 29 dB,while a 30.4 dB power budget is obtained in the 50 Gbit/s experimental transmission.展开更多
The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle...The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle remains a challenging task.To tackle this challenge,the present study proposes a novel approach for identifying the gradient-distributed plastic parameters for the S38C axle by integrating nano-indentation techniques with the machine learning method.Firstly,nano-indentation tests are conducted along the radial direction of the S38C axle to obtain the gradient-distributed load-displacement curves,nano-hardness,and elastic modulus.Subsequently,the dimensionless analysis is performed to obtain the representative stress,strain,and yield stress from load-displacement curves.These parameters are then incorporated into the machine learning method as physical information to identify the gradient-distributed plastic parameters of the S38C axle.The results indicate that the proposed method based on the physics-informed neural network and multi-fidelity neural network successfully identifies the gradient-distributed plastic parameters of the S38C axles and demonstrates superior prediction accuracy and generalization compared with the purely data-driven machine learning method.展开更多
Mango is a plant with high economic value in the agricultural industry;thus,it is necessary to maximize the productivity performance of the mango plant,which can be done by implementing artificial intelligence.In this...Mango is a plant with high economic value in the agricultural industry;thus,it is necessary to maximize the productivity performance of the mango plant,which can be done by implementing artificial intelligence.In this study,a lightweight object detection model will be developed that can detect mango plant conditions based on disease potential,so that it becomes an early detection warning system that has an impact on increasing agricultural productivity.The proposed lightweight model integrates YOLOv7-Tiny and the proposed modules,namely the C2S module.The C2S module consists of three sub-modules such as the convolutional block attention module(CBAM),the coordinate attention(CA)module,and the squeeze-and-excitation(SE)module.The dataset is constructed by eight classes,including seven classes of disease conditions and one class of health conditions.The experimental result shows that the proposed lightweight model has the optimal results,which increase by 13.15% of mAP50 compared to the original model YOLOv7-Tiny.While the mAP50:95 also achieved the highest results compared to other models,including YOLOv3-Tiny,YOLOv4-Tiny,YOLOv5,and YOLOv7-Tiny.The advantage of the proposed lightweightmodel is the adaptability that supports it in constrained environments,such as edge computing systems.This proposedmodel can support a robust,precise,and convenient precision agriculture system for the user.展开更多
Since the United Nations launched the Sustainable Development Goals(SDGs)in 2015,global implementation has steadily advanced,yet prominent challenges persist.Progress has been uneven across regions and countries,with ...Since the United Nations launched the Sustainable Development Goals(SDGs)in 2015,global implementation has steadily advanced,yet prominent challenges persist.Progress has been uneven across regions and countries,with Tajikistan representing a typical example of such disparities.Based on 81 SDG indicators for Tajikistan from 2001 to 2023,this study applied a three-level coupling network framework:at the microscale,it identified synergies and trade-offs between indicators;at the mesoscale,it examined the strength and direction of linkages within four SDG-related components(society,finance,governance,and environment);and at the global level,it focused on the overall SDG interlinkages.Spearman’s rank correlation,sliding window method,and topological properties were employed to analyze the coupling dynamics of SDGs.Results showed that over 70.00%of associations in the global SDG network were of medium-to-low intensity,alongside extremely strong ones(|r|value approached 1.00,where r is the correlation coefficient).SDG interactions were generally limited,with stable local synergy clusters in core livelihood sectors.Network modularity fluctuated,reflecting a cycle of differentiation,integration,and fragmentation,while coupling efficiency varied with the external environment.Each component exhibited distinct functional characteristics.The social component maintained high connectivity through the“poverty alleviation-education-healthcare”loop.The environmental component shifted toward coordinated eco-economic governance.The governance-related component broke interdepartmental barriers,while the financial component showed weak links between resource-based indicators and consumption/employment indicators.Tajikistan’s SDG coupling evolved through three phases:survival-oriented(2001–2012),policy integration(2013–2018),and shock adaptation(2019–2023).These phases were driven by policy changes,resource industries,governance optimization,and external factors.This study enriches the analytical framework for understanding the dynamic coupling of SDGs in mountainous resource-dependent countries and provides empirical evidence to support similar countries in formulating phase-specific SDG promotion strategies.展开更多
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
文摘Objective Repetitive transcranial magnetic stimulation(rTMS)has demonstrated efficacy in enhancing neurocognitive performance in Alzheimer’s disease(AD),but the neurobiological mechanisms linking synaptic pathology,neural oscillatory dynamics,and brain network reorganization remain unclear.This investigation seeks to systematically evaluate the therapeutic potential of rTMS as a non-invasive neuromodulatory intervention through a multimodal framework integrating clinical assessments,molecular profiling,and neurophysiological monitoring.Methods In this prospective double-blind trial,12 AD patients underwent a 14-day protocol of 20 Hz rTMS,with comprehensive multimodal assessments performed pre-and postintervention.Cognitive functioning was quantified using the mini-mental state examination(MMSE)and Montreal cognitive assessment(MOCA),while daily living capacities and neuropsychiatric profiles were respectively evaluated through the activities of daily living(ADL)scale and combined neuropsychiatric inventory(NPI)-Hamilton depression rating scale(HAMD).Peripheral blood biomarkers,specifically Aβ1-40 and phosphorylated tau(p-tau181),were analyzed to investigate the effects of rTMS on molecular metabolism.Spectral power analysis was employed to investigate rTMS-induced modulations of neural rhythms in AD patients,while brain network analyses incorporating topological properties were conducted to examine stimulus-driven network reorganization.Furthermore,systematic assessment of correlations between cognitive scale scores,blood biomarkers,and network characteristics was performed to elucidate cross-modal therapeutic associations.Results Clinically,MMSE and MOCA scores improved significantly(P<0.05).Biomarker showed that Aβ1-40 level increased(P<0.05),contrasting with p-tau181 reduction.Moreover,the levels of Aβ1-40 were positively correlated with MMSE and MOCA scores.Post-intervention analyses revealed significant modulations in oscillatory power,characterized by pronounced reductions in delta(P<0.05)and theta bands(P<0.05),while concurrent enhancements were observed in alpha,beta,and gamma band activities(all P<0.05).Network analysis revealed frequency-specific reorganization:clustering coefficients were significantly decreased in delta,theta,and alpha bands(P<0.05),while global efficiency improvement was exclusively detected in the delta band(P<0.05).The alpha band demonstrated concurrent increases in average nodal degree(P<0.05)and characteristic path length reduction(P<0.05).Further research findings indicate that the changes in the clinical scale HAMD scores before and after rTMS stimulation are negatively correlated with the changes in the blood biomarkers Aβ1-40 and p-tau181.Additionally,the changes in the clinical scales MMSE and MoCA scores were negatively correlated with the changes in the node degree of the alpha frequency band and negatively correlated with the clustering coefficient of the delta frequency band.However,the changes in MMSE scores are positively correlated with the changes in global efficiency of both the delta and alpha frequency bands.Conclusion 20 Hz rTMS targeting dorsolateral prefrontal cortex(DLPFC)significantly improves cognitive function and enhances the metabolic clearance ofβ-amyloid and tau proteins in AD patients.This neurotherapeutic effect is mechanistically associated with rTMS-mediated frequency-selective neuromodulation,which enhances the connectivity of oscillatory networks through improved neuronal synchronization and optimized topological organization of functional brain networks.These findings not only support the efficacy of rTMS as an adjunctive therapy for AD but also underscore the importance of employing multiple assessment methods—including clinical scales,blood biomarkers,and EEG——in understanding and monitoring the progression of AD.This research provides a significant theoretical foundation and empirical evidence for further exploration of rTMS applications in AD treatment.
基金funded by Zhejiang Province Traditional Chinese Medicine Science and Technology Program(No.2021ZZ012)The Changlin Qiu National Distinguished Senior Traditional Chinese Medicine Expert Heritage Workshop Project(No.GZS2021007).
文摘Background:Parkinson’s disease(PD)is one of the most common movement disorders worldwide.Ziyin Xifeng Decoction(ZYXFD),a traditional Chinese medicine compound formula,has shown therapeutic efficacy in treating PD,but its specific mechanisms of action have not been fully elucidated.Methods:Firstly,we employed network pharmacology and untargeted metabolomics analysis to identify the core targets,pathways,and key metabolites of ZYXFD in the treatment of PD.Subsequently,we evaluated the protective effects of ZYXFD and further investigated its anti-PD mechanisms by validating the analytical results.Results:Combined analyses of network pharmacology and metabolomics identify the core targets including EGFR,SRC,PTGS2,and CDK2,while the effects of ZYXFD against PD are likely mediated primarily through the PI3K/AKT/mTOR signaling pathway.Pharmacodynamic evaluation demonstrated that a high dose of ZYXFD significantly improved behavioral deficits in chronic PD mice,downregulatedα-synuclein protein expression,and protected dopaminergic neurons.It also regulated the expression of core targets,inhibited the PI3K/AKT/mTOR signaling pathway,promoted autophagy,and reduced apoptosis.In vitro experiments further verified that the therapeutic effect of ZYXFD on PD is dependent on autophagy regulation.Conclusion:The findings demonstrated that ZYXFD alleviates PD by modulating related proteins and metabolites,inhibiting the PI3K/AKT/mTOR signaling pathway,and enhancing autophagy.This provides a theoretical basis for its broader application in PD treatment.
基金supported by the Shenzhen Fundamental Research Program,No.JCYJ20240813151132042the National Natural Science Foundation of China,Nos.32270701 and 32470708+1 种基金Young Talent Recruitment Project of Guangdong,No.2019QN01Y139the Science and Technology Planning Project of Guangdong Province,No.2023B1212060018(all to GL).
文摘Mitochondrial DNA variants have been linked to cognitive progression in Parkinson’s disease;however,the mechanisms by which mitochondrial DNA variants or haplogroups contribute to this process remain unclear.In the present study,we analyzed single-nucleus RNA sequencing data from 241 post-mortem brain samples across five regions to investigate the dysregulatory mechanisms associated with mitochondrial DNA haplogroup H and haplogroups J,T,and U#.Our findings revealed significant alterations in the proportions of astrocyte subtypes CHI3L1 and GRM3 in the neocortical regions of haplogroup H.Notably,TTR was markedly downregulated in the dorsal motor nucleus of the Xth nerve region of patients with haplogroup H.Pathway analysis highlighted abnormal hypoxic and reactive oxygen species environments in astrocytes,whereas protein complex analysis revealed a consistent and significant elevation in ribosomal subunit complexes within the astrocyte subtypes.By constructing weighted and directed transcriptome-wide gene regulatory networks,we identified significant changes in transcription factor SP1 and homeobox protein HOXA5 activity in the astrocyte subtypes of individuals with haplogroup H.Additionally,widespread dysregulation was observed in the transcriptional control of TTR by multiple transcription factors.Parkinson’s disease patients with haplogroup H also exhibited increased network functional connectivity in specific brain regions.This data-driven study underscores the potential mechanisms by which mitochondrial DNA haplogroups contribute to cognitive progression in Parkinson’s disease,involving cellular composition changes,differential gene expression,pathway disruption,and gene regulatory networks.Our findings suggest that mitochondrial DNA haplogroup H may drive Parkinson’s disease cognitive progression through aberrant TTR expression and a hypoxic environment.
基金ZTE Industry-University-Institute Cooperation Funds under Grant No.HC-CN-20230105001National Natural Science Foundation of China under Grant No.62001045。
文摘The 200 Gbit/s passive optical network(PON)is most likely to be the next-generation scheme following 50G PON.The costeffective direct detection(DD)system is the economical choice.However,larger-capacity DD systems will face much more serious power fading caused by chromatic dispersion(CD)combined with square-law DD and thereby significantly increases the complexity of equalization algorithms.In this paper,a 200 Gbit/s Nyquist 4-level pulse amplitude modulation(PAM4)single side-band(SSB)modulation-DD downlink scheme is designed,and a low complexity quadratic-nonlinear equalizer is proposed for this system.The computational complexity of the quadratic nonlinear equalizer is about 28%of that of the conventional Volterra nonlinear equalizer,while still exhibiting excellent nonlinear equalization ability.Simulation results for the 200 Gbit/s system with 20 km fiber transmission show that it can achieve a power budget of 29 dB,while a 30.4 dB power budget is obtained in the 50 Gbit/s experimental transmission.
基金supported by the National Key Research and Development Plan(Grant No.2022YFB3401901)the National Natural Science Foundation of China(Grant Nos.12192210,12192214,12072295,and 12222209)+1 种基金Independent Project of State Key Laboratory of Rail Transit Vehicle System(Grant No.2023TPL-T03)Fundamental Research Funds for the Central Universities(Grant No.2682023CG004).
文摘The S38C railway axle undergoes induction hardening,resulting in a gradient-distributed microstructure and mechanical properties.The accurate identification of gradient-distributed plastic parameters for the S38C axle remains a challenging task.To tackle this challenge,the present study proposes a novel approach for identifying the gradient-distributed plastic parameters for the S38C axle by integrating nano-indentation techniques with the machine learning method.Firstly,nano-indentation tests are conducted along the radial direction of the S38C axle to obtain the gradient-distributed load-displacement curves,nano-hardness,and elastic modulus.Subsequently,the dimensionless analysis is performed to obtain the representative stress,strain,and yield stress from load-displacement curves.These parameters are then incorporated into the machine learning method as physical information to identify the gradient-distributed plastic parameters of the S38C axle.The results indicate that the proposed method based on the physics-informed neural network and multi-fidelity neural network successfully identifies the gradient-distributed plastic parameters of the S38C axles and demonstrates superior prediction accuracy and generalization compared with the purely data-driven machine learning method.
基金supported by National Science and Technology Council(NSTC)Taiwan,Grant No.NSTC 113-2221-E-167-023.
文摘Mango is a plant with high economic value in the agricultural industry;thus,it is necessary to maximize the productivity performance of the mango plant,which can be done by implementing artificial intelligence.In this study,a lightweight object detection model will be developed that can detect mango plant conditions based on disease potential,so that it becomes an early detection warning system that has an impact on increasing agricultural productivity.The proposed lightweight model integrates YOLOv7-Tiny and the proposed modules,namely the C2S module.The C2S module consists of three sub-modules such as the convolutional block attention module(CBAM),the coordinate attention(CA)module,and the squeeze-and-excitation(SE)module.The dataset is constructed by eight classes,including seven classes of disease conditions and one class of health conditions.The experimental result shows that the proposed lightweight model has the optimal results,which increase by 13.15% of mAP50 compared to the original model YOLOv7-Tiny.While the mAP50:95 also achieved the highest results compared to other models,including YOLOv3-Tiny,YOLOv4-Tiny,YOLOv5,and YOLOv7-Tiny.The advantage of the proposed lightweightmodel is the adaptability that supports it in constrained environments,such as edge computing systems.This proposedmodel can support a robust,precise,and convenient precision agriculture system for the user.
文摘Since the United Nations launched the Sustainable Development Goals(SDGs)in 2015,global implementation has steadily advanced,yet prominent challenges persist.Progress has been uneven across regions and countries,with Tajikistan representing a typical example of such disparities.Based on 81 SDG indicators for Tajikistan from 2001 to 2023,this study applied a three-level coupling network framework:at the microscale,it identified synergies and trade-offs between indicators;at the mesoscale,it examined the strength and direction of linkages within four SDG-related components(society,finance,governance,and environment);and at the global level,it focused on the overall SDG interlinkages.Spearman’s rank correlation,sliding window method,and topological properties were employed to analyze the coupling dynamics of SDGs.Results showed that over 70.00%of associations in the global SDG network were of medium-to-low intensity,alongside extremely strong ones(|r|value approached 1.00,where r is the correlation coefficient).SDG interactions were generally limited,with stable local synergy clusters in core livelihood sectors.Network modularity fluctuated,reflecting a cycle of differentiation,integration,and fragmentation,while coupling efficiency varied with the external environment.Each component exhibited distinct functional characteristics.The social component maintained high connectivity through the“poverty alleviation-education-healthcare”loop.The environmental component shifted toward coordinated eco-economic governance.The governance-related component broke interdepartmental barriers,while the financial component showed weak links between resource-based indicators and consumption/employment indicators.Tajikistan’s SDG coupling evolved through three phases:survival-oriented(2001–2012),policy integration(2013–2018),and shock adaptation(2019–2023).These phases were driven by policy changes,resource industries,governance optimization,and external factors.This study enriches the analytical framework for understanding the dynamic coupling of SDGs in mountainous resource-dependent countries and provides empirical evidence to support similar countries in formulating phase-specific SDG promotion strategies.
