期刊文献+
共找到9篇文章
< 1 >
每页显示 20 50 100
Deep Convolutional Neural Networks for Accurate Classification of Gastrointestinal Tract Syndromes
1
作者 Zahid Farooq Khan Muhammad Ramzan +4 位作者 Mudassar Raza Muhammad Attique Khan Khalid Iqbal Taerang Kim Jae-Hyuk Cha 《Computers, Materials & Continua》 SCIE EI 2024年第1期1207-1225,共19页
Accurate detection and classification of artifacts within the gastrointestinal(GI)tract frames remain a significant challenge in medical image processing.Medical science combined with artificial intelligence is advanc... Accurate detection and classification of artifacts within the gastrointestinal(GI)tract frames remain a significant challenge in medical image processing.Medical science combined with artificial intelligence is advancing to automate the diagnosis and treatment of numerous diseases.Key to this is the development of robust algorithms for image classification and detection,crucial in designing sophisticated systems for diagnosis and treatment.This study makes a small contribution to endoscopic image classification.The proposed approach involves multiple operations,including extracting deep features from endoscopy images using pre-trained neural networks such as Darknet-53 and Xception.Additionally,feature optimization utilizes the binary dragonfly algorithm(BDA),with the fusion of the obtained feature vectors.The fused feature set is input into the ensemble subspace k nearest neighbors(ESKNN)classifier.The Kvasir-V2 benchmark dataset,and the COMSATS University Islamabad(CUI)Wah private dataset,featuring three classes of endoscopic stomach images were used.Performance assessments considered various feature selection techniques,including genetic algorithm(GA),particle swarm optimization(PSO),salp swarm algorithm(SSA),sine cosine algorithm(SCA),and grey wolf optimizer(GWO).The proposed model excels,achieving an overall classification accuracy of 98.25% on the Kvasir-V2 benchmark and 99.90% on the CUI Wah private dataset.This approach holds promise for developing an automated computer-aided system for classifying GI tract syndromes through endoscopy images. 展开更多
关键词 feature fusion Darknet-53 Xception binary dragonfly algorithm ENSEMBLE
下载PDF
新53阶次月球重力场及其精细结构
2
作者 杨宏伟 赵文津 《地球物理学报》 SCIE EI CAS CSCD 北大核心 2012年第7期2268-2280,共13页
月球重力场是了解月球内部结构的重要信息之一.日本SELENE卫星首次获得月球背面卫星轨道的直接探测数据并建立了更高精度的全月球重力场模型.本文根据日本公布的采样间隔为60s、轨道高度为100km的SELENE卫星观测资料并利用作者移植的GEO... 月球重力场是了解月球内部结构的重要信息之一.日本SELENE卫星首次获得月球背面卫星轨道的直接探测数据并建立了更高精度的全月球重力场模型.本文根据日本公布的采样间隔为60s、轨道高度为100km的SELENE卫星观测资料并利用作者移植的GEODYN-II微机版本软件求解出新53阶次月球球谐场模型LG-53.经过测试表明移植后的微机版本比原始工作站版本的计算效率提高了5到10倍.理论上表明60s采样间隔、100km高度的轨道数据能够计算出60阶次的月球球谐系数模型,但是作者在实际计算过程中发现:在接近理论阶次(60阶次)的一系列模型中出现了平行于经线的高频噪声,且模型越接近理论阶次其噪声越高.因此本文将53阶次月球球谐系数模型LG-53作为最后的解算结果并建立各种月球重力异常场,并将其与美国GLGM-2(70阶次)模型和利用嫦娥1号数据解算出的CEGM-01(50阶次)模型对比,发现新53阶次重力场模型LG-53在高纬度和月球背面都显示出了更高分辨率的异常特征;与美国LP165P(165阶次)模型对比发现LG-53所建立的自由空气重力异常在月球背面不存在LP165P中所出现的高频噪声.与日本90阶次SGM90d模型对比后发现新模型的精度较日本模型还有所差距.主要是由于两者参与计算的数据采样率不同所致.53阶次的模型LG-53能够反映100km尺度的重力异常,而日本90阶次模型则可以反映60km尺度的异常.利用新53阶次模型计算的自由空气重力异常图并结合月球地形图探讨了四种类型的Mascon重力异常特征及其地形特征. 展开更多
关键词 LG-53 SELENE轨道数据 GEODYN-II软件 Mascon重力异常特征
下载PDF
Myc诱导的核抗原53在结直肠癌中的表达及其与病理特征的关系
3
作者 司华 闫灿 曾宪焕 《实用癌症杂志》 2021年第4期574-577,共4页
目的探讨Myc诱导的核抗原53(mina53)在结直肠癌中的表达及其与病理特征的关系。方法选择初治的结直肠癌患者341例作为研究对象,取所有患者的病灶组织标本(病灶组)与癌旁正常结直肠黏膜组织(癌旁组),采用免疫组化法检测mina53表达水平,... 目的探讨Myc诱导的核抗原53(mina53)在结直肠癌中的表达及其与病理特征的关系。方法选择初治的结直肠癌患者341例作为研究对象,取所有患者的病灶组织标本(病灶组)与癌旁正常结直肠黏膜组织(癌旁组),采用免疫组化法检测mina53表达水平,调查患者的病理特征、预后并进行相关性分析。结果病灶组mina53表达阳性率为55.1%,显著高于癌旁组的19.6%(P<0.05)。mina53表达阳性与结直肠癌患者年龄、性别、部位、直径无相关性(P>0.05),与分化程度、临床分期、浆膜浸润有相关性(P<0.05)。所有患者均随访到2020年7月1日,随访死亡48例,死亡率14.1%。多因素Cox比例风险回归分析显示:分化程度、临床分期、浆膜浸润、mina53表达阳性为导致患者预后死亡的主要因素(P<0.05)。结论mina53在结直肠癌中呈现高表达状况,与患者的分化程度、临床分期、浆膜浸润等病理特征显著相关,也是影响患者预后的重要因素。 展开更多
关键词 Myc诱导的核抗原53 结直肠癌中 病理特征 预后 相关性
下载PDF
食管粘膜下鳞癌生物学特性及P^(53)蛋白表达的意义
4
作者 张三申 韩孝存 +5 位作者 郭梅 张云汉 赵子贞 肖郑生 张立新 杨怀华 《中国肿瘤临床与康复》 2000年第2期24-25,共2页
目的 研究食管粘膜下鳞癌的生物学特性。方法 通过随访及SP免疫组化技术P53 蛋白在 3 0例食管粘膜下鳞癌的表达 ,研究其生物学特性。结果  3 0例食管粘膜下鳞癌的P53 蛋白阳性表达率为 66.7% ,经病理证实术时淋巴结的转移率为 60 % ... 目的 研究食管粘膜下鳞癌的生物学特性。方法 通过随访及SP免疫组化技术P53 蛋白在 3 0例食管粘膜下鳞癌的表达 ,研究其生物学特性。结果  3 0例食管粘膜下鳞癌的P53 蛋白阳性表达率为 66.7% ,经病理证实术时淋巴结的转移率为 60 % ,术后各种转移率为 47.8% ,其中伴肺转移 4例 ,单纯肺转移 1例 ,其 1、2、3、4年生存率分别为 10 0 %、70 .0 %、3 1.3 %和 11.1%。结论 食管粘膜下鳞癌术时及术后转移率较高 ,术后肺转移率较高 ,P53 蛋白阳性表达率高 ,予后差。鳞化伴异型增生的食管腺上皮细胞的阳性表达提示食管粘膜下鳞癌可能发生于食管腺。 展开更多
关键词 食管粘膜下鳞状细胞癌 P^(53)蛋白 免疫组化 生物学特性
下载PDF
融合定位算法的YOLO-V4模型实现电力杆塔状态评估
5
作者 阮远峰 郭建武 蔡金涛 《电工技术》 2023年第13期32-34,40,共4页
针对电网线路在使用中经常出现断杆、倒杆现象,而现有的技术性检测方式已不能满足要求的问题,给出了一种新的电力杆塔状态评估方法,并建立了改善的YOLO-V4模型,以实现电力杆塔的状态精准定位。CSPDarknet-53模型试验说明,该方法能精准... 针对电网线路在使用中经常出现断杆、倒杆现象,而现有的技术性检测方式已不能满足要求的问题,给出了一种新的电力杆塔状态评估方法,并建立了改善的YOLO-V4模型,以实现电力杆塔的状态精准定位。CSPDarknet-53模型试验说明,该方法能精准定位电力杆塔故障点,偏差小。 展开更多
关键词 电网线路 状态评估 YOLO-V4模型 cspdarknet-53特征 目标检测
下载PDF
基于YOLO V4的机房异常巡检研究 被引量:1
6
作者 曾路 汪浩 孙骏 《电力大数据》 2022年第6期56-61,共6页
为实时检测电网系统中机房异常情况,避免产生安全隐患和财产损失,提出了一种基于深度学习目标检测理论的机房异常巡检模型。该方法以CSPDarkNet-53卷积模型作为骨干网络,在数据预处理环节引入多种数据增强手段弥补数据数量不足的局限性... 为实时检测电网系统中机房异常情况,避免产生安全隐患和财产损失,提出了一种基于深度学习目标检测理论的机房异常巡检模型。该方法以CSPDarkNet-53卷积模型作为骨干网络,在数据预处理环节引入多种数据增强手段弥补数据数量不足的局限性,防止产生过拟合现象。以电网系统某机房为对象进行实验。实验结果表明,该方法能有效地检测到机房中常见的异常状况,检测精度符合实际场景需求。 展开更多
关键词 异常情况 YOLO V4 cspdarknet-53 骨干网络 数据增强
下载PDF
改进型YOLO-V4模型的电力杆塔状态评估探索 被引量:2
7
作者 张宝星 毕明利 张壮领 《信息技术》 2021年第8期81-86,91,共7页
针对电网线路存在倒杆、断杆现象以及现有技术巡检方式落后的问题,提出一种新型的电力杆塔状态评估方法,构建出能够实现电力杆塔位置定位的评估方法,并引入电力杆塔检测的YOLO-V4模型深度学习算法,该算法模型包括53个卷积层,具有大量的... 针对电网线路存在倒杆、断杆现象以及现有技术巡检方式落后的问题,提出一种新型的电力杆塔状态评估方法,构建出能够实现电力杆塔位置定位的评估方法,并引入电力杆塔检测的YOLO-V4模型深度学习算法,该算法模型包括53个卷积层,具有大量的3*3、1*1的卷积核,该算法还具有Darknet-53特征提取网络、多尺度融合特征网等,通过评价函数对所应用的YOLO-V4目标检测网络的损失进行检测。试验表明,YOLO-V4模型深度学习算法引入GIoU指标后,相比普通状况平均精度(AP)从97.12%提高到98.94%,准确率(Precision)从94.5%提高至95.6%,召回率(Recall)从97.5%提高至99.2%。 展开更多
关键词 电网线路 状态评估 YOLO-V4模型 Darknet-53特征 目标检测
下载PDF
P53基因蛋白质序列的相似性及其聚类分析 被引量:2
8
作者 仇建烨 朱平 《计算机与应用化学》 CAS CSCD 北大核心 2013年第9期1043-1046,共4页
基于详细HP模型,先把20种氨基酸分为四类。再建立一个连续的坐标空间,对蛋白质序列进行混沌游走(CGR),把蛋白质序列中的每个氨基酸映射到平面坐标系中,每个氨基酸对应着一个点。然后用四类氨基酸的坐标值的平均值组成8维特征向量对蛋白... 基于详细HP模型,先把20种氨基酸分为四类。再建立一个连续的坐标空间,对蛋白质序列进行混沌游走(CGR),把蛋白质序列中的每个氨基酸映射到平面坐标系中,每个氨基酸对应着一个点。然后用四类氨基酸的坐标值的平均值组成8维特征向量对蛋白质序列进行数值刻画,并定义向量间欧式距离为序列间的距离。最后基于这种对蛋白质序列的数学描述,对包括人类在内的8个物种的p53基因蛋白质序列进行了相似性分析和模糊聚类。结果与实际相符,表明了用此方法研究蛋白质序列相似的合理性。这对研究蛋白质序列有着重要意义。 展开更多
关键词 蛋白质 混沌游走 特征向量 P53 模糊聚类
原文传递
Etiological characteristics and treatment of tardive dyskinesia
9
作者 Zhe Li Xueli Sun Che Zhou 《Neural Regeneration Research》 SCIE CAS CSCD 2006年第6期549-552,共4页
OBJECTIVE : The pathogenesis of tardive dyskinesia (TD) is complicated and uncertain, Thus, there is not any effective treatment for it. The psychiatrists pay more and more attention to TD, which lasts for a long t... OBJECTIVE : The pathogenesis of tardive dyskinesia (TD) is complicated and uncertain, Thus, there is not any effective treatment for it. The psychiatrists pay more and more attention to TD, which lasts for a long time and is difficult to treat. DATA SOURCES: A computer-based online search of Medline database was undertaken to identify articles about the feature of etiology and the progression of treatment for TD published in English by using the keywords of "rD, etiology, pathogenesis" and "TD, therapy, drug treatment". Meanwhile, Chinese articles about the feature of etiology and the progression of treatment for TD were searched in Wanfang database and China journal full-text database, and the keywords were "TD, etiology, pathogenesis" and "TD, therapy, drug Treatment" in Chinese. STUDY SELECTION: Articles met the following inclusion criteria were selected in this paper. Inclusion criteria: (1) Researches of randomized blind control design, before and after control design and retrospective. (2) Researches of the feature of etiology and the progression of treatment for TD. Exclusion criteria: the repetitive researches and individual reports. DATA EXTRACTION : Totally 65 articles related the feature of etiology and the progression of treatment for TD of randomized blind control design, before and after control design and retrospective studies were collected, and 53 of them were accorded with the inclusion criteria. Of the 12 excluded ones, 8 were concerning with genetics, 4 were repetitive researches. DATA SYNTHESIS : The feature of etiology for TD includes:(1) Hypothesis of dopamine receptor super-sensitivity: The dopamine receptor is persistently blocked, so it will result in functional disorder in CNS, and then TD may take place. (2)) Hypothesis of neuronal degeneration: The concentration of aminosuccinic acid and glutamic acid will increase after the antipsychotic used for a long time and this will result in neuronal degeneration through glutamic acid receptor in the postsynaptic membrane; meanwhile with free radical, the nerve cells of corpus striatum may degenerate and become necrosis. (3) Sex and age: The females and gerontal patients are liability to the TD disease. It is may related to the lower estrogen. (4) Molecule heredity: TD may association with the dopamine and 5-HT receptor gene polymorphism. (5) Other theories: Hypofunction of γ-amino-butyri acid (GABA), hypothesis of noradrenaline 5-serotonin and nutrition metabolism can cause TD disease. Treatlent for TD: (1) Dopamine receptor agonist: The therapeutic effect is not satisfactory, especially for gerontism females. (2) Oxygen free radical scavenger: As represent of vitamin E, it can clear out free radicals and reduce the potential cytotoxic effect of free radicals. (3) Calcium channel blocker: This maybe related to block calcium ions releasing from muscle cells and inhibit muscle convulsion; therefore, it can be used for symptomatic treatment. (4) GABA receptor agonist: It is more effective for the prominent dysmyotonia than dancing slowly symptom. (5) Antipsychotic: There is some therapeutic effect with ciozapine, but the effect will reduce because of the age growing up and the symptom exacerbating. (6) Other therapies: Valproate sodium, cyproheptadine, melatonin, branched chain amino acid, ahalysantinfarctasum, electric acupuncture and injection ad acumen, traditional Chinese drug have a certain effects on TD. Prevention of TD: The serum creatine phosphokinase (CPK) combined with symptoms should be checked regularly so as to early discovery TD. CONCLUSION : (1) Etiology of TD: The hypothesis of dopamine receptor super-sensitivity is denyed; the hypothesis of neuronal degeneration is approved in academic circles; the sex and age is a finding of generally received; but the dopamine and 5-HT receptor gene polymorphism, hypofunction of GABA, noradrenaline, 5-serotonin and nutrition metabolism cannot explain the pathogenesis of TD. (2) Treatment for TD: The therapeutic effect of dopamine receptor agonist is not satisfactory; the oxygen free radical scavenger maybe effective; calcium channel blocker maybe used for symptomatic treatment; GABA receptor agonist maybe more effective for the prominent dysmyotonia than dancing slowly symptom; the consequence of antipsychotic is discrepancy; other therapies maybe use to adjunctive therapies. (3) As far as prevention of TD is concerned, and the serum CPK combined with symptoms should be checked regularly so as to early discovery TD. 展开更多
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部