期刊文献+
共找到4篇文章
< 1 >
每页显示 20 50 100
S型曲线的扫描回归方法 被引量:4
1
作者 孙耀东 王太源 陶俊 《扬州大学学报(自然科学版)》 CAS CSCD 1999年第3期10-12,共3页
讨论S型曲线Y=k[1+bexp(-aX)]的最佳拟合问题,提出一种扫描回归的新方法,并利用实际数据说明,与普通回归相比,该扫描回归方法具有相关系数较大。
关键词 线性回归 扫描回归 S型曲线 最佳拟合
下载PDF
全基因组混合模型关联分析的极速回归扫描法研究 被引量:2
2
作者 赵敬丽 李淑玲 +1 位作者 高进 杨润清 《东北农业大学学报》 CAS CSCD 北大核心 2018年第7期58-66,共9页
在全基因组混合模型关联分析(GWMMAS)中,利用剩余多基因遗传力代替方差比值(剩余多基因方差/误差方差),将多基因遗传力求解限制在(0,1)区间内。引入R语言Rcpp Armadillo程序包中的极速线性模型拟合函数(fast Lm Pure函数)快速估计单核... 在全基因组混合模型关联分析(GWMMAS)中,利用剩余多基因遗传力代替方差比值(剩余多基因方差/误差方差),将多基因遗传力求解限制在(0,1)区间内。引入R语言Rcpp Armadillo程序包中的极速线性模型拟合函数(fast Lm Pure函数)快速估计单核苷酸多态性(SNP)效应和完整LMM最大似然值。从由GBLUP估计的性状基因组遗传力出发,逐个高通量SNP的GWMMAS约需4次全基因组回归扫描。当仅关注EMMAX法估计的大效应或高显著水准标记时,GWMMAS运行时间缩短在两次扫描之内。与采用lm函数优化剩余多基因方差比的Fa ST-LMM法相比,极速回归扫描法可成倍提高GWMMAS计算效率。计算机模拟试验证实新方法统计和计算效率,运用极速回归扫描法可高效定位与牙鲆生长性状相关基因位点。 展开更多
关键词 全基因组混合模型关联分析 极速线性模型拟合函数 微效多基因遗传力 最大似然估计 基因组回归扫描
下载PDF
可线性化回归方法的改进和拓展 被引量:4
3
作者 孙耀东 王太源 宗序平 《扬州大学学报(自然科学版)》 CAS CSCD 2001年第2期11-14,共4页
改进了可线性化回归方法的理论 ,给出了可线性化回归新方法 ,利用实例说明 ,与传统回归法相比 ,新方法具有偏差平方和小、精确度高的优点 .并在新方法的基础上进一步拓展 ,提出扫描回归方法 ,解决一类非线性回归问题 .与马阔特 ( Marqua... 改进了可线性化回归方法的理论 ,给出了可线性化回归新方法 ,利用实例说明 ,与传统回归法相比 ,新方法具有偏差平方和小、精确度高的优点 .并在新方法的基础上进一步拓展 ,提出扫描回归方法 ,解决一类非线性回归问题 .与马阔特 ( Marquardt)方法相比 ,所得回归曲线更理想 。 展开更多
关键词 可线性化 偏差平方和 扫描回归方法
下载PDF
New scoring system for intra-abdominal injury diagno- sis after blunt trauma 被引量:1
4
作者 Majid Sho.Jaee Gholamreza Faridaalaee +4 位作者 Mahmoud Yousefifard Mehdi Yaseri Aft Arhami Dolatabadi AnitaSabzghabaei Ali Malekirastekenari 《Chinese Journal of Traumatology》 CAS CSCD 2014年第1期19-24,共6页
Objective: An accurate scoring system for intra-abdominal injury (IAI) based on clinical manifesta- tion and examination may decrease unnecessary CT scans, save time, and reduce healthcare cost. This study is de- s... Objective: An accurate scoring system for intra-abdominal injury (IAI) based on clinical manifesta- tion and examination may decrease unnecessary CT scans, save time, and reduce healthcare cost. This study is de- signed to provide a new scoring system for a better diagno- sis of IAI after blunt trauma. Methods: This prospective observational study was performed from April 2011 to October 2012 on patients aged above 18 years and suspected with blunt abdominal trauma (BAT) admitted to the emergency department (ED) of Imam Hussein Hospital and Shohadaye Hafte Tir Hospital. All patients were assessed and treated based on Advanced Trauma Life Support and ED protocol. Diagnosis was done according to CT scan findings, which was considered as the gold standard. Data were gathered based on patient's history, physical exam, ultrasound and CT scan findings by a general practitioner who was not blind to this study. Chi- square test and logistic regression were done. Factors with significant relationship with CT scan were imported in multi- variate regression models, where a coefficient (13) was given based on the contribution of each of them. Scoring systemwas developed based on the obtained total [3 of each factor. Results: Altogether 261 patients (80.1% male) were enrolled (48 cases of IAI). A 24-point blunt abdominal trauma scoring system (BATSS) was developed. Patients were di- vided into three groups including low (score〈8), moderate (8≤score〈12) and high risk (score≥ 12). In high risk group immediate laparotomy should be done, moderate group needs further assessments, and low risk group should be kept under observation. Low risk patients did not show positive CT-scans (specificity 100%). Conversely, all high risk patients had positive CT-scan findings (sensitivity 100%). The receiver operating characteristic curve indicated a close relationship between the results of CT scan and BATSS (sensitivity=99.3%). Conclusion: The present scoring system furnishes a high precision and reproducible diagnostic tool for BAT detection and has the potential to reduce unnecessary CT scan and cut unnecessary costs. 展开更多
关键词 Abdominal injuries Tomography X-ray computed DIAGNOSIS
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部