Artificial intelligence(AI)is increasingly recognized as a transformative force in the field of solid organ transplantation.From enhancing donor-recipient matching to predicting clinical risks and tailoring immunosupp...Artificial intelligence(AI)is increasingly recognized as a transformative force in the field of solid organ transplantation.From enhancing donor-recipient matching to predicting clinical risks and tailoring immunosuppressive therapy,AI has the potential to improve both operational efficiency and patient outcomes.Despite these advancements,the perspectives of transplant professionals-those at the forefront of critical decision-making-remain insufficiently explored.To address this gap,this study utilizes a multi-round electronic Delphi approach to gather and analyses insights from global experts involved in organ transplantation.Participants are invited to complete structured surveys capturing demographic data,professional roles,institutional practices,and prior exposure to AI technologies.The survey also explores perceptions of AI’s potential benefits.Quantitative responses are analyzed using descriptive statistics,while open-ended qualitative responses undergo thematic analysis.Preliminary findings indicate a generally positive outlook on AI’s role in enhancing transplantation processes,particularly in areas such as donor matching and post-operative care.These mixed views reflect both optimism and caution among professionals tasked with integrating new technologies into high-stakes clinical workflows.By capturing a wide range of expert opinions,the findings will inform future policy development,regulatory considerations,and institutional readiness frameworks for the integration of AI into organ transplantation.展开更多
Contrastive graph clustering(CGC)has become a prominent method for self-supervised representation learning by contrasting augmented graph data pairs.However,the performance of CGC methods critically depends on the cho...Contrastive graph clustering(CGC)has become a prominent method for self-supervised representation learning by contrasting augmented graph data pairs.However,the performance of CGC methods critically depends on the choice of data augmentation,which usually limits the capacity of network generalization.Besides,most existing methods characterize positive and negative samples based on the nodes themselves,ignoring the influence of neighbors with different hop numbers on the node.In this study,a novel self-cumulative contrastive graph clustering(SC-CGC)method is devised,which is capable of dynamically adjusting the influence of neighbors with different hops.Our intuition is that better neighbors are closer and distant ones are further away in their feature space,thus we can perform neighbor contrasting without data augmentation.To be specific,SC-CGC relies on two neural networks,i.e.,autoencoder network(AE)and graph autoencoder network(GAE),to encode the node information and graph structure,respectively.To make these two networks interact and learn from each other,a dynamic fusion mechanism is devised to transfer the knowledge learned by AE to the corresponding GAE layer by layer.Then,a self-cumulative contrastive loss function is designed to characterize the structural information by dynamically accumulating the influence of the nodes with different hops.Finally,our approach simultaneously refines the representation learning and clustering assignments in a self-supervised manner.Extensive experiments on 8 realistic datasets demonstrate that SC-CGC consistently performs better over SOTA techniques.The code is available at https://github.com/Xiaoqiang-Yan/JAS-SCCGC.展开更多
The successful penetration of government,corporate,and organizational IT systems by state and non-state actors deploying APT vectors continues at an alarming pace.Advanced Persistent Threat(APT)attacks continue to pos...The successful penetration of government,corporate,and organizational IT systems by state and non-state actors deploying APT vectors continues at an alarming pace.Advanced Persistent Threat(APT)attacks continue to pose significant challenges for organizations despite technological advancements in artificial intelligence(AI)-based defense mechanisms.While AI has enhanced organizational capabilities for deterrence,detection,and mitigation of APTs,the global escalation in reported incidents,particularly those successfully penetrating critical government infrastructure has heightened concerns among information technology(IT)security administrators and decision-makers.Literature review has identified the stealthy lateral movement(LM)of malware within the initially infected local area network(LAN)as a significant concern.However,current literature has yet to propose a viable approach for resource-efficient,real-time detection of APT malware lateral movement within the initially compromised LAN following perimeter breach.Researchers have suggested the nature of the dataset,optimal feature selection,and the choice of machine learning(ML)techniques as critical factors for detection.Hence,the objective of the research described here was to successfully demonstrate a simplified lightweight ML method for detecting the LM of APT vectors.While the nearest detection rate achieved in the LM domain within LAN was 99.89%,as reported in relevant studies,our approach surpassed it,with a detection rate of 99.95%for the modified random forest(RF)classifier for dataset 1.Additionally,our approach achieved a perfect 100%detection rate for the decision tree(DT)and RF classifiers with dataset 2,a milestone not previously reached in studies within this domain involving two distinct datasets.Using the ML life cycle methodology,we deployed K-nearest neighbor(KNN),support vector machine(SVM),DT,and RF on three relevant datasets to detect the LM of APTs at the affected LAN prior to data exfiltration/destruction.Feature engineering presented four critical APT LM intrusion detection(ID)indicators(features)across the three datasets,namely,the source port number,the destination port number,the packets,and the bytes.This study demonstrates the effectiveness of lightweight ML classifiers in detecting APT lateral movement after network perimeter breach.It contributes to the field by proposing a non-intrusive network detection method capable of identifying APT malware before data exfiltration,thus providing an additional layer of organizational defense.展开更多
Designing fast and accurate neural networks is becoming essential in various vision tasks.Recently,the use of attention mechanisms has increased,aimed at enhancing the vision task performance by selectively focusing o...Designing fast and accurate neural networks is becoming essential in various vision tasks.Recently,the use of attention mechanisms has increased,aimed at enhancing the vision task performance by selectively focusing on relevant parts of the input.In this paper,we concentrate on squeeze-and-excitation(SE)-based channel attention,considering the trade-off between latency and accuracy.We propose a variation of the SE module,called squeeze-and-excitation with layer normalization(SELN),in which layer normalization(LN)replaces the sigmoid activation function.This approach reduces the vanishing gradient problem while enhancing feature diversity and discriminability of channel attention.In addition,we propose a latency-efficient model named SELNeXt,where the LN typically used in the ConvNext block is replaced by SELN to minimize additional latency-impacting operations.Through classification simulations on ImageNet-1k,we show that the top-1 accuracy of the proposed SELNeXt outperforms other ConvNeXt-based models in terms of latency efficiency.SELNeXt also achieves better object detection and instance segmentation performance on COCO than Swin Transformer and ConvNeXt for small-sized models.Our results indicate that LN could be a considerable candidate for replacing the activation function in attention mechanisms.In addition,SELNeXt achieves a better accuracy-latency trade-off,making it favorable for real-time applications and edge computing.The code is available at https://github.com/oto-q/SELNeXt(accessed on 06 December 2024).展开更多
Task-oriented point cloud sampling aims to select a representative subset from the input,tailored to specific application scenarios and task requirements.However,existing approaches rarely tackle the problem of redund...Task-oriented point cloud sampling aims to select a representative subset from the input,tailored to specific application scenarios and task requirements.However,existing approaches rarely tackle the problem of redundancy caused by local structural similarities in 3D objects,which limits the performance of sampling.To address this issue,this paper introduces a novel task-oriented point cloud masked autoencoder-based sampling network(Point-MASNet),inspired by the masked autoencoder mechanism.Point-MASNet employs a voxel-based random non-overlapping masking strategy,which allows the model to selectively learn and capture distinctive local structural features from the input data.This approach effectively mitigates redundancy and enhances the representativeness of the sampled subset.In addition,we propose a lightweight,symmetrically structured keypoint reconstruction network,designed as an autoencoder.This network is optimized to efficiently extract latent features while enabling refined reconstructions.Extensive experiments demonstrate that Point-MASNet achieves competitive sampling performance across classification,registration,and reconstruction tasks.展开更多
Paediatric liver transplantation(PLT)is a life-saving procedure for children with advanced liver disease or hepatoblastoma.The number of available grafts is limited in relation to the number of children on PLT waiting...Paediatric liver transplantation(PLT)is a life-saving procedure for children with advanced liver disease or hepatoblastoma.The number of available grafts is limited in relation to the number of children on PLT waiting list.This graft shortage has led transplant societies and healthcare organizations to explore ways to investigate possible options and expand the donor pool.The safe use of grafts from obese donors has always been a subject of debate among PLT specialists.Donors’obesity is strongly associated with hepatic steatosis which can affect graft function by impairing microcirculation and maximizing the potential of ischemiareperfusion injury.Donor body mass index consideration should go hand in hand with the workup for hepatic steatosis which is an independent predictor for early graft dysfunction.New strategies to optimize the grafts before PLT such as normothermic regional perfusion and ex vivo liver perfusion can potentially mitigate the risk of using grafts from obese donors.This review summarizes the available evidence about the impact of donor obesity on PLT and highlights the current policies to widen the graft pool and suggest future research directions to improve donor selection and patient outcomes.展开更多
AIM: To establish the potential of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBHHx) as a material for tendon repair. METHODS: The biocompatibility of PHBHHx with both rat tenocytes (rT) and human mesenchymal ste...AIM: To establish the potential of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBHHx) as a material for tendon repair. METHODS: The biocompatibility of PHBHHx with both rat tenocytes (rT) and human mesenchymal stem cells (hMSC) was explored by monitoring adhesive characteristics on films of varying weight/volume ratios coupled to a culture atmosphere of either 21% O2 (air) or 2% O2 (physiological normoxia). The diameter and stiffness of PHBHHx films was established using optical coherence tomography and mechanical testing, respectively. RESULTS: Film thickness correlated directly with weight/volume PHBHHx (r2 = 0.9473) ranging from 0.1 mm (0.8% weight/volume) to 0.19 mm (2.4% weight/volume). Film stiffness on the other hand displayed a biphasic response which increased rapidly at values > 1.6% weight/volume. Optimal cell attachment of rT required films of ≥ 1.6% and ≥ 2.0% weight/volume PHBHHx in 2% O2 and 21% O2 respectively. A qualitative adhesion increase was noted for hMSC in films ≥ 1.2% weight/volume, becoming significant at 2% weight/volume in 2% O2. An increase in cell adhesion was also noted with ≥ 2% weight/volume PHBHHx in 21% O2. Cell migration into films was not observed. CONCLUSION: This evaluation demonstrates that PHBHHx is a suitable polymer for future cell/polymer replacement strategies in tendon repair.展开更多
文摘Artificial intelligence(AI)is increasingly recognized as a transformative force in the field of solid organ transplantation.From enhancing donor-recipient matching to predicting clinical risks and tailoring immunosuppressive therapy,AI has the potential to improve both operational efficiency and patient outcomes.Despite these advancements,the perspectives of transplant professionals-those at the forefront of critical decision-making-remain insufficiently explored.To address this gap,this study utilizes a multi-round electronic Delphi approach to gather and analyses insights from global experts involved in organ transplantation.Participants are invited to complete structured surveys capturing demographic data,professional roles,institutional practices,and prior exposure to AI technologies.The survey also explores perceptions of AI’s potential benefits.Quantitative responses are analyzed using descriptive statistics,while open-ended qualitative responses undergo thematic analysis.Preliminary findings indicate a generally positive outlook on AI’s role in enhancing transplantation processes,particularly in areas such as donor matching and post-operative care.These mixed views reflect both optimism and caution among professionals tasked with integrating new technologies into high-stakes clinical workflows.By capturing a wide range of expert opinions,the findings will inform future policy development,regulatory considerations,and institutional readiness frameworks for the integration of AI into organ transplantation.
基金supported by the National Natural Science Foundation of China(62371423,62450002,62425107)China Postdoctoral Science Foundation(2020M682357).
文摘Contrastive graph clustering(CGC)has become a prominent method for self-supervised representation learning by contrasting augmented graph data pairs.However,the performance of CGC methods critically depends on the choice of data augmentation,which usually limits the capacity of network generalization.Besides,most existing methods characterize positive and negative samples based on the nodes themselves,ignoring the influence of neighbors with different hop numbers on the node.In this study,a novel self-cumulative contrastive graph clustering(SC-CGC)method is devised,which is capable of dynamically adjusting the influence of neighbors with different hops.Our intuition is that better neighbors are closer and distant ones are further away in their feature space,thus we can perform neighbor contrasting without data augmentation.To be specific,SC-CGC relies on two neural networks,i.e.,autoencoder network(AE)and graph autoencoder network(GAE),to encode the node information and graph structure,respectively.To make these two networks interact and learn from each other,a dynamic fusion mechanism is devised to transfer the knowledge learned by AE to the corresponding GAE layer by layer.Then,a self-cumulative contrastive loss function is designed to characterize the structural information by dynamically accumulating the influence of the nodes with different hops.Finally,our approach simultaneously refines the representation learning and clustering assignments in a self-supervised manner.Extensive experiments on 8 realistic datasets demonstrate that SC-CGC consistently performs better over SOTA techniques.The code is available at https://github.com/Xiaoqiang-Yan/JAS-SCCGC.
基金Rabdan Academy for funding the research presented in the paper.
文摘The successful penetration of government,corporate,and organizational IT systems by state and non-state actors deploying APT vectors continues at an alarming pace.Advanced Persistent Threat(APT)attacks continue to pose significant challenges for organizations despite technological advancements in artificial intelligence(AI)-based defense mechanisms.While AI has enhanced organizational capabilities for deterrence,detection,and mitigation of APTs,the global escalation in reported incidents,particularly those successfully penetrating critical government infrastructure has heightened concerns among information technology(IT)security administrators and decision-makers.Literature review has identified the stealthy lateral movement(LM)of malware within the initially infected local area network(LAN)as a significant concern.However,current literature has yet to propose a viable approach for resource-efficient,real-time detection of APT malware lateral movement within the initially compromised LAN following perimeter breach.Researchers have suggested the nature of the dataset,optimal feature selection,and the choice of machine learning(ML)techniques as critical factors for detection.Hence,the objective of the research described here was to successfully demonstrate a simplified lightweight ML method for detecting the LM of APT vectors.While the nearest detection rate achieved in the LM domain within LAN was 99.89%,as reported in relevant studies,our approach surpassed it,with a detection rate of 99.95%for the modified random forest(RF)classifier for dataset 1.Additionally,our approach achieved a perfect 100%detection rate for the decision tree(DT)and RF classifiers with dataset 2,a milestone not previously reached in studies within this domain involving two distinct datasets.Using the ML life cycle methodology,we deployed K-nearest neighbor(KNN),support vector machine(SVM),DT,and RF on three relevant datasets to detect the LM of APTs at the affected LAN prior to data exfiltration/destruction.Feature engineering presented four critical APT LM intrusion detection(ID)indicators(features)across the three datasets,namely,the source port number,the destination port number,the packets,and the bytes.This study demonstrates the effectiveness of lightweight ML classifiers in detecting APT lateral movement after network perimeter breach.It contributes to the field by proposing a non-intrusive network detection method capable of identifying APT malware before data exfiltration,thus providing an additional layer of organizational defense.
基金supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education under Grant NRF-2021R1A6A1A03039493.
文摘Designing fast and accurate neural networks is becoming essential in various vision tasks.Recently,the use of attention mechanisms has increased,aimed at enhancing the vision task performance by selectively focusing on relevant parts of the input.In this paper,we concentrate on squeeze-and-excitation(SE)-based channel attention,considering the trade-off between latency and accuracy.We propose a variation of the SE module,called squeeze-and-excitation with layer normalization(SELN),in which layer normalization(LN)replaces the sigmoid activation function.This approach reduces the vanishing gradient problem while enhancing feature diversity and discriminability of channel attention.In addition,we propose a latency-efficient model named SELNeXt,where the LN typically used in the ConvNext block is replaced by SELN to minimize additional latency-impacting operations.Through classification simulations on ImageNet-1k,we show that the top-1 accuracy of the proposed SELNeXt outperforms other ConvNeXt-based models in terms of latency efficiency.SELNeXt also achieves better object detection and instance segmentation performance on COCO than Swin Transformer and ConvNeXt for small-sized models.Our results indicate that LN could be a considerable candidate for replacing the activation function in attention mechanisms.In addition,SELNeXt achieves a better accuracy-latency trade-off,making it favorable for real-time applications and edge computing.The code is available at https://github.com/oto-q/SELNeXt(accessed on 06 December 2024).
基金supported by the National Key Research and Development Program of China(2022YFB3103500)the National Natural Science Foundation of China(62473033,62571027)+1 种基金in part by the Beijing Natural Science Foundation(L231012)the State Scholarship Fund from the China Scholarship Council.
文摘Task-oriented point cloud sampling aims to select a representative subset from the input,tailored to specific application scenarios and task requirements.However,existing approaches rarely tackle the problem of redundancy caused by local structural similarities in 3D objects,which limits the performance of sampling.To address this issue,this paper introduces a novel task-oriented point cloud masked autoencoder-based sampling network(Point-MASNet),inspired by the masked autoencoder mechanism.Point-MASNet employs a voxel-based random non-overlapping masking strategy,which allows the model to selectively learn and capture distinctive local structural features from the input data.This approach effectively mitigates redundancy and enhances the representativeness of the sampled subset.In addition,we propose a lightweight,symmetrically structured keypoint reconstruction network,designed as an autoencoder.This network is optimized to efficiently extract latent features while enabling refined reconstructions.Extensive experiments demonstrate that Point-MASNet achieves competitive sampling performance across classification,registration,and reconstruction tasks.
文摘Paediatric liver transplantation(PLT)is a life-saving procedure for children with advanced liver disease or hepatoblastoma.The number of available grafts is limited in relation to the number of children on PLT waiting list.This graft shortage has led transplant societies and healthcare organizations to explore ways to investigate possible options and expand the donor pool.The safe use of grafts from obese donors has always been a subject of debate among PLT specialists.Donors’obesity is strongly associated with hepatic steatosis which can affect graft function by impairing microcirculation and maximizing the potential of ischemiareperfusion injury.Donor body mass index consideration should go hand in hand with the workup for hepatic steatosis which is an independent predictor for early graft dysfunction.New strategies to optimize the grafts before PLT such as normothermic regional perfusion and ex vivo liver perfusion can potentially mitigate the risk of using grafts from obese donors.This review summarizes the available evidence about the impact of donor obesity on PLT and highlights the current policies to widen the graft pool and suggest future research directions to improve donor selection and patient outcomes.
基金Supported by EPSRC Doctoral Training Centre in Regenerative Medicine and the HYANJI Scaffold Project (European Commission Framework 7 program)
文摘AIM: To establish the potential of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBHHx) as a material for tendon repair. METHODS: The biocompatibility of PHBHHx with both rat tenocytes (rT) and human mesenchymal stem cells (hMSC) was explored by monitoring adhesive characteristics on films of varying weight/volume ratios coupled to a culture atmosphere of either 21% O2 (air) or 2% O2 (physiological normoxia). The diameter and stiffness of PHBHHx films was established using optical coherence tomography and mechanical testing, respectively. RESULTS: Film thickness correlated directly with weight/volume PHBHHx (r2 = 0.9473) ranging from 0.1 mm (0.8% weight/volume) to 0.19 mm (2.4% weight/volume). Film stiffness on the other hand displayed a biphasic response which increased rapidly at values > 1.6% weight/volume. Optimal cell attachment of rT required films of ≥ 1.6% and ≥ 2.0% weight/volume PHBHHx in 2% O2 and 21% O2 respectively. A qualitative adhesion increase was noted for hMSC in films ≥ 1.2% weight/volume, becoming significant at 2% weight/volume in 2% O2. An increase in cell adhesion was also noted with ≥ 2% weight/volume PHBHHx in 21% O2. Cell migration into films was not observed. CONCLUSION: This evaluation demonstrates that PHBHHx is a suitable polymer for future cell/polymer replacement strategies in tendon repair.
