Lung cancer continues to be a leading cause of cancer-related deaths worldwide,emphasizing the critical need for improved diagnostic techniques.Early detection of lung tumors significantly increases the chances of suc...Lung cancer continues to be a leading cause of cancer-related deaths worldwide,emphasizing the critical need for improved diagnostic techniques.Early detection of lung tumors significantly increases the chances of successful treatment and survival.However,current diagnostic methods often fail to detect tumors at an early stage or to accurately pinpoint their location within the lung tissue.Single-model deep learning technologies for lung cancer detection,while beneficial,cannot capture the full range of features present in medical imaging data,leading to incomplete or inaccurate detection.Furthermore,it may not be robust enough to handle the wide variability in medical images due to different imaging conditions,patient anatomy,and tumor characteristics.To overcome these disadvantages,dual-model or multi-model approaches can be employed.This research focuses on enhancing the detection of lung cancer by utilizing a combination of two learning models:a Convolutional Neural Network(CNN)for categorization and the You Only Look Once(YOLOv8)architecture for real-time identification and pinpointing of tumors.CNNs automatically learn to extract hierarchical features from raw image data,capturing patterns such as edges,textures,and complex structures that are crucial for identifying lung cancer.YOLOv8 incorporates multiscale feature extraction,enabling the detection of tumors of varying sizes and scales within a single image.This is particularly beneficial for identifying small or irregularly shaped tumors that may be challenging to detect.Furthermore,through the utilization of cutting-edge data augmentation methods,such as Deep Convolutional Generative Adversarial Networks(DCGAN),the suggested approach can handle the issue of limited data and boost the models’ability to learn from diverse and comprehensive datasets.The combined method not only improved accuracy and localization but also ensured efficient real-time processing,which is crucial for practical clinical applications.The CNN achieved an accuracy of 97.67%in classifying lung tissues into healthy and cancerous categories.The YOLOv8 model achieved an Intersection over Union(IoU)score of 0.85 for tumor localization,reflecting high precision in detecting and marking tumor boundaries within the images.Finally,the incorporation of synthetic images generated by DCGAN led to a 10%improvement in both the CNN classification accuracy and YOLOv8 detection performance.展开更多
In this research, one-dimensional stratiform a novel dual-model system, cold cloud model （1DSC） coupled to Weather Research and Forecast （WRF） model （WRF-1DSC for short）, was employed to investigate the effects ...In this research, one-dimensional stratiform a novel dual-model system, cold cloud model （1DSC） coupled to Weather Research and Forecast （WRF） model （WRF-1DSC for short）, was employed to investigate the effects of cloud seeding by silver iodide （AgI） on rain enhancement. Driven by changing environmental conditions extracted from the WRF model, WRF-1DSC could be used to assess the cloud seeding effects quantitatively. The employment of WRF- 1DSC, in place of a one-dimen- sional two-moment cloud seeding model applied to a three-dimensional mesoscale cloud-resolving model, was found to result in massive reduction of computational resources. Numerical experiments with WRF-1DSC were conducted for a real stratiform precipitation event ob- served on 4-5 July 2004, in Northeast China. A good agreement between the observed and modeled cloud system ensured the ability of WRF-1DSC to simulate the observed precipitation process efficiently. Sensitivity tests were performed with different seeding times, locations, and amounts. Experimental results showed that the optimum seeding effect （defined as the percentage of rain enhancement or rain enhancement rate） could be achieved through proper seeding at locations of maximum cloud water content when the updraft was strong. The optimum seeding effect was found to increase by 5.61% when the cloud was seeded at 5.5 km above ground level around 2300 UTC 4 July 2004, with the maximum AgI mixing ratio （As） equaling 15 ng kg-1. On the other hand, for an overseeded cloud, a significant reduction occurred in the accumulated precipitation （-12.42%） as Xs reached 100 ng kg^-1. This study demonstrates the potential of WRF- 1DSC in determining the optimal AgI seeding strategy in practical operations of precipitation enhancement.展开更多
It is urgent to find a technology accurately to better diagnose and treat to brain tumor.Eu-doped Gd2 O3 nanorods(Eu-Gd2 O3 NRs)with paramagnetic and fluorescent properties were conjugated with doxorubicin(Dox)and chl...It is urgent to find a technology accurately to better diagnose and treat to brain tumor.Eu-doped Gd2 O3 nanorods(Eu-Gd2 O3 NRs)with paramagnetic and fluorescent properties were conjugated with doxorubicin(Dox)and chlorotoxin(CTX)via PEGylation,hydrazone bond and sulfur bond(named as CTXNRs-Dox),and these NRs could release more Dox in lower pH environment.The results of cell experiments indicated that CTX-NRs-Dox had obvious targeting and toxic effects on U251 cells,as well as good fluorescence imaging behavior.The orthotopic glioma-transplanted mice models were constructed via the intracranial injection of glioma cells(U87 MG).The result of experiments after the tail-vein injection of the prepared NRs suggested that CTX-NRs-Dox could target to brain tumors via the long-time blood circulation,leading to their obvious contrast enhancement of MR imaging of the intracranial tumor and their significant inhibitory effect on the growth and metastasis of brain tumors.A mechanism of synergistic effect of CTX-NRs-Dox on targeting and inhabiting the brain tumor was proposed.Our research suggested that CTX-NRs-Dox had potential application prospect in the detection and treatment of glioma.展开更多
时间序列预测在能源管理、交通流量和气象分析等多个实际场景中具有重要应用价值。然而,时间序列数据中存在的分布漂移(Distribution Shift)与长程依赖(Long-term Dependency)仍限制了传统方法与现有深度学习模型在长期预测中的表现。为...时间序列预测在能源管理、交通流量和气象分析等多个实际场景中具有重要应用价值。然而,时间序列数据中存在的分布漂移(Distribution Shift)与长程依赖(Long-term Dependency)仍限制了传统方法与现有深度学习模型在长期预测中的表现。为此,提出了一种名为D-LINet(Dual-Normalization and Linear Integration Network)的创新模型。该模型结合了Dish-TS(Distribution Shift in Time Series Forecasting)框架的分布归一化能力与线性映射的高效性,并采用双向归一化与双线性层的设计,有效缓解输入与输出空间的分布偏移,增强了对周期性与趋势性特征的捕捉能力。在多个真实数据集上对D-LINet的预测性能进行了全面评估。结果显示,在短期与长期预测中,D-LINet的均方误差和平均绝对误差均显著优于主流模型(如Transformer,Informer,Autoformer和DLinear)。此外,实验还探讨了输入窗口长度及先验知识的引入对预测性能的影响,为后续模型优化提供了重要指导。该研究针对复杂分布漂移问题提出了新的解决思路,并有助于提升时间序列预测的精度与稳健性。展开更多
文摘Lung cancer continues to be a leading cause of cancer-related deaths worldwide,emphasizing the critical need for improved diagnostic techniques.Early detection of lung tumors significantly increases the chances of successful treatment and survival.However,current diagnostic methods often fail to detect tumors at an early stage or to accurately pinpoint their location within the lung tissue.Single-model deep learning technologies for lung cancer detection,while beneficial,cannot capture the full range of features present in medical imaging data,leading to incomplete or inaccurate detection.Furthermore,it may not be robust enough to handle the wide variability in medical images due to different imaging conditions,patient anatomy,and tumor characteristics.To overcome these disadvantages,dual-model or multi-model approaches can be employed.This research focuses on enhancing the detection of lung cancer by utilizing a combination of two learning models:a Convolutional Neural Network(CNN)for categorization and the You Only Look Once(YOLOv8)architecture for real-time identification and pinpointing of tumors.CNNs automatically learn to extract hierarchical features from raw image data,capturing patterns such as edges,textures,and complex structures that are crucial for identifying lung cancer.YOLOv8 incorporates multiscale feature extraction,enabling the detection of tumors of varying sizes and scales within a single image.This is particularly beneficial for identifying small or irregularly shaped tumors that may be challenging to detect.Furthermore,through the utilization of cutting-edge data augmentation methods,such as Deep Convolutional Generative Adversarial Networks(DCGAN),the suggested approach can handle the issue of limited data and boost the models’ability to learn from diverse and comprehensive datasets.The combined method not only improved accuracy and localization but also ensured efficient real-time processing,which is crucial for practical clinical applications.The CNN achieved an accuracy of 97.67%in classifying lung tissues into healthy and cancerous categories.The YOLOv8 model achieved an Intersection over Union(IoU)score of 0.85 for tumor localization,reflecting high precision in detecting and marking tumor boundaries within the images.Finally,the incorporation of synthetic images generated by DCGAN led to a 10%improvement in both the CNN classification accuracy and YOLOv8 detection performance.
基金supported by the Knowledge Innovation Program of Chinese Academy of Sciences (Grant No. KZCX2-EW-203)the National Basic Research Program of China (Grant No.2013CB430105)the National Department Public Benefit Research Foundation (Grant No.GYHY201006031)
文摘In this research, one-dimensional stratiform a novel dual-model system, cold cloud model （1DSC） coupled to Weather Research and Forecast （WRF） model （WRF-1DSC for short）, was employed to investigate the effects of cloud seeding by silver iodide （AgI） on rain enhancement. Driven by changing environmental conditions extracted from the WRF model, WRF-1DSC could be used to assess the cloud seeding effects quantitatively. The employment of WRF- 1DSC, in place of a one-dimen- sional two-moment cloud seeding model applied to a three-dimensional mesoscale cloud-resolving model, was found to result in massive reduction of computational resources. Numerical experiments with WRF-1DSC were conducted for a real stratiform precipitation event ob- served on 4-5 July 2004, in Northeast China. A good agreement between the observed and modeled cloud system ensured the ability of WRF-1DSC to simulate the observed precipitation process efficiently. Sensitivity tests were performed with different seeding times, locations, and amounts. Experimental results showed that the optimum seeding effect （defined as the percentage of rain enhancement or rain enhancement rate） could be achieved through proper seeding at locations of maximum cloud water content when the updraft was strong. The optimum seeding effect was found to increase by 5.61% when the cloud was seeded at 5.5 km above ground level around 2300 UTC 4 July 2004, with the maximum AgI mixing ratio （As） equaling 15 ng kg-1. On the other hand, for an overseeded cloud, a significant reduction occurred in the accumulated precipitation （-12.42%） as Xs reached 100 ng kg^-1. This study demonstrates the potential of WRF- 1DSC in determining the optimal AgI seeding strategy in practical operations of precipitation enhancement.
基金supported by the National Natural Science Foundation of China (Nos.51273122,51872190)Sichuan Science and Technology Project (No.2018JY0535)supported by the Fundamental of Research Funds for the Central University (Nos.SCU2017A001,2018SCUH0024)
文摘It is urgent to find a technology accurately to better diagnose and treat to brain tumor.Eu-doped Gd2 O3 nanorods(Eu-Gd2 O3 NRs)with paramagnetic and fluorescent properties were conjugated with doxorubicin(Dox)and chlorotoxin(CTX)via PEGylation,hydrazone bond and sulfur bond(named as CTXNRs-Dox),and these NRs could release more Dox in lower pH environment.The results of cell experiments indicated that CTX-NRs-Dox had obvious targeting and toxic effects on U251 cells,as well as good fluorescence imaging behavior.The orthotopic glioma-transplanted mice models were constructed via the intracranial injection of glioma cells(U87 MG).The result of experiments after the tail-vein injection of the prepared NRs suggested that CTX-NRs-Dox could target to brain tumors via the long-time blood circulation,leading to their obvious contrast enhancement of MR imaging of the intracranial tumor and their significant inhibitory effect on the growth and metastasis of brain tumors.A mechanism of synergistic effect of CTX-NRs-Dox on targeting and inhabiting the brain tumor was proposed.Our research suggested that CTX-NRs-Dox had potential application prospect in the detection and treatment of glioma.
文摘时间序列预测在能源管理、交通流量和气象分析等多个实际场景中具有重要应用价值。然而,时间序列数据中存在的分布漂移(Distribution Shift)与长程依赖(Long-term Dependency)仍限制了传统方法与现有深度学习模型在长期预测中的表现。为此,提出了一种名为D-LINet(Dual-Normalization and Linear Integration Network)的创新模型。该模型结合了Dish-TS(Distribution Shift in Time Series Forecasting)框架的分布归一化能力与线性映射的高效性,并采用双向归一化与双线性层的设计,有效缓解输入与输出空间的分布偏移,增强了对周期性与趋势性特征的捕捉能力。在多个真实数据集上对D-LINet的预测性能进行了全面评估。结果显示,在短期与长期预测中,D-LINet的均方误差和平均绝对误差均显著优于主流模型(如Transformer,Informer,Autoformer和DLinear)。此外,实验还探讨了输入窗口长度及先验知识的引入对预测性能的影响,为后续模型优化提供了重要指导。该研究针对复杂分布漂移问题提出了新的解决思路,并有助于提升时间序列预测的精度与稳健性。
