a****g 发帖数: 8131 | |
h*****o 发帖数: 47 | 2 medical device 的作用一般比较local,drug除了对target的部位有影响,也可能对其
它部位有影响。这样导致medical device的数据exploratory的部分比较少,常规性的
分析比较多,所以medical device 公司,统计师相对少一些。
just my 2 cents. |
a****g 发帖数: 8131 | 3 thanks
好像是这样子
【在 h*****o 的大作中提到】 : medical device 的作用一般比较local,drug除了对target的部位有影响,也可能对其 : 它部位有影响。这样导致medical device的数据exploratory的部分比较少,常规性的 : 分析比较多,所以medical device 公司,统计师相对少一些。 : just my 2 cents.
|
c****g 发帖数: 156 | 4 The trial design is very different for treatment device and diagnostic
device too. |
D*********2 发帖数: 535 | 5 那是不是会偏工程一点,reliability啊design of experiment啊神马的比较多? |
c****g 发帖数: 156 | 6 A lot of clinical evaluations for diagnostic device have two parts:
imprecision (reproducibility) and clinical performance (sensitivity/
specificity etc.). The former involves DOE and ANOVA etc, the later uses ROC
, regression, categorical analysis etc. |
a****g 发帖数: 8131 | 7 what is the full name of ROC? thanks.
ROC
【在 c****g 的大作中提到】 : A lot of clinical evaluations for diagnostic device have two parts: : imprecision (reproducibility) and clinical performance (sensitivity/ : specificity etc.). The former involves DOE and ANOVA etc, the later uses ROC : , regression, categorical analysis etc.
|
Y******Y 发帖数: 8753 | 8 receiver operating characteristic
【在 a****g 的大作中提到】 : what is the full name of ROC? thanks. : : ROC
|
D*********2 发帖数: 535 | 9 Thanks.
ROC
【在 c****g 的大作中提到】 : A lot of clinical evaluations for diagnostic device have two parts: : imprecision (reproducibility) and clinical performance (sensitivity/ : specificity etc.). The former involves DOE and ANOVA etc, the later uses ROC : , regression, categorical analysis etc.
|