针对传统机器学习方法依赖人工特征提取,存在检测算法准确率低、无法应对0day漏洞利用等未知类型攻击等问题,提出一种基于卷积神经网络(Convolutional Neural Networks,CNN)和长短期记忆网络(Long-Short Term Memory,LSTM)混合算法的异...针对传统机器学习方法依赖人工特征提取,存在检测算法准确率低、无法应对0day漏洞利用等未知类型攻击等问题,提出一种基于卷积神经网络(Convolutional Neural Networks,CNN)和长短期记忆网络(Long-Short Term Memory,LSTM)混合算法的异常流量检测方法,充分发掘攻击流量的结构化特点,提取流量数据的时空特征,提高了异常流量检测系统性能。实验结果表明,在CIC-IDS2017数据集上,多种异常流量检测的准确率均超过96.9%,总体准确率达到98.8%,与其他机器学习算法相比准确率更高,同时保持了极低的误警率。展开更多
Background Non-small cell lung cancer(NSCLC)patients treated with immune checkpoint inhibitors(ICIs)are at an increased risk of cardiovascular adverse events during surgery.This study investigated the impact of periop...Background Non-small cell lung cancer(NSCLC)patients treated with immune checkpoint inhibitors(ICIs)are at an increased risk of cardiovascular adverse events during surgery.This study investigated the impact of perioperative sufentanil administration on reducing these risks.Methods From July 2021 to July 2022,this observational study included 139 NSCLC patients at our institution,split into an experimental group(n=88)receiving sufentanil during surgery and a control group(n=51)who did not.The study focused on evaluating major adverse cardiovascular events(MACEs),arrhythmias,elevated troponin I,and B-type natriuretic peptide(BNP)levels as primary endpoints.Results The experimental group experienced a lower incidence of arrhythmias at 9.1%compared to 19.6%in the control group(P=0.045),indicating a significant protective effect of sufentanil.No significant differences were found in MACEs rates(8.0%in the experimental group vs.11.8%in the control group,P=0.710),elevated troponin I(12.5%vs.19.6%,P=0.147),or BNP levels(26.1%vs.31.4%,P=0.875).Patients with existing cardiovascular risk factors demonstrated a higher overall incidence of cardiovascular adverse events,underscoring the compounded risk that these factors posed.Conclusions Perioperative administration of sufentanil in NSCLC patients undergoing surgery post-ICI therapy shows a significant reduction in the incidence of arrhythmias,suggesting a protective cardiovascular effect.This finding highlighted the importance of sufentanil in managing the perioperative cardiovascular risk of these patients.[S Chin J Cardiol 2024;25(1):21-28].展开更多
DNS滥用已成为网络空间安全治理中面临的最具挑战性的威胁之一。针对现有检测方法多以DNS请求流量为研究对象,忽略了响应流量特征的问题,提出一种基于长短期记忆网络(Long-Short Term Memory,LSTM)的DNS隐蔽信道检测方法。综合分析请求...DNS滥用已成为网络空间安全治理中面临的最具挑战性的威胁之一。针对现有检测方法多以DNS请求流量为研究对象,忽略了响应流量特征的问题,提出一种基于长短期记忆网络(Long-Short Term Memory,LSTM)的DNS隐蔽信道检测方法。综合分析请求与响应流量特征,提取响应流量中时间戳、TTL、响应分组长度等特征点,并构建LSTM模型进行训练。实验结果表明,该方法在准确率、F1评分等指标方面取得了良好的结果,较现有方法有显著提高。展开更多
文摘针对传统机器学习方法依赖人工特征提取,存在检测算法准确率低、无法应对0day漏洞利用等未知类型攻击等问题,提出一种基于卷积神经网络(Convolutional Neural Networks,CNN)和长短期记忆网络(Long-Short Term Memory,LSTM)混合算法的异常流量检测方法,充分发掘攻击流量的结构化特点,提取流量数据的时空特征,提高了异常流量检测系统性能。实验结果表明,在CIC-IDS2017数据集上,多种异常流量检测的准确率均超过96.9%,总体准确率达到98.8%,与其他机器学习算法相比准确率更高,同时保持了极低的误警率。
基金supported by the Guangzhou City Science and Technology Plan Project(No.202102080438)。
文摘Background Non-small cell lung cancer(NSCLC)patients treated with immune checkpoint inhibitors(ICIs)are at an increased risk of cardiovascular adverse events during surgery.This study investigated the impact of perioperative sufentanil administration on reducing these risks.Methods From July 2021 to July 2022,this observational study included 139 NSCLC patients at our institution,split into an experimental group(n=88)receiving sufentanil during surgery and a control group(n=51)who did not.The study focused on evaluating major adverse cardiovascular events(MACEs),arrhythmias,elevated troponin I,and B-type natriuretic peptide(BNP)levels as primary endpoints.Results The experimental group experienced a lower incidence of arrhythmias at 9.1%compared to 19.6%in the control group(P=0.045),indicating a significant protective effect of sufentanil.No significant differences were found in MACEs rates(8.0%in the experimental group vs.11.8%in the control group,P=0.710),elevated troponin I(12.5%vs.19.6%,P=0.147),or BNP levels(26.1%vs.31.4%,P=0.875).Patients with existing cardiovascular risk factors demonstrated a higher overall incidence of cardiovascular adverse events,underscoring the compounded risk that these factors posed.Conclusions Perioperative administration of sufentanil in NSCLC patients undergoing surgery post-ICI therapy shows a significant reduction in the incidence of arrhythmias,suggesting a protective cardiovascular effect.This finding highlighted the importance of sufentanil in managing the perioperative cardiovascular risk of these patients.[S Chin J Cardiol 2024;25(1):21-28].
文摘DNS滥用已成为网络空间安全治理中面临的最具挑战性的威胁之一。针对现有检测方法多以DNS请求流量为研究对象,忽略了响应流量特征的问题,提出一种基于长短期记忆网络(Long-Short Term Memory,LSTM)的DNS隐蔽信道检测方法。综合分析请求与响应流量特征,提取响应流量中时间戳、TTL、响应分组长度等特征点,并构建LSTM模型进行训练。实验结果表明,该方法在准确率、F1评分等指标方面取得了良好的结果,较现有方法有显著提高。
