临床预测模型中连续变量的降级处理方式

2021-04-11 杨弘 BioMed数据坊

在临床预测模型的构建中,为了让模型有更好的适用性和简便性,许多研究人员将连续型自变量转换为分类变量。

在临床预测模型的构建中,为了让模型有更好的适用性和简便性,许多研究人员将连续型自变量转换为分类变量。 降级处理的首要原因是为了在临床实践中“分类”的变量简单判别,使预测模型的结果更实用;其次,clinical prediction rules(CPRs)中分类变量广泛应用,在模型整体表现最优的基础上,一个模型的适用性取决于变量是否简单易判。 从纯统计的角度看,将连续变量降级处理为分类变量会导致信息的丢失与模型预测能力的降低。但是从临床角度看,如果预测因子是分类的,更容易快速的对患者进行结局的判定。此外,当前多数预测模型都假设Y与X之间是线性关系,但是若两者不服从该假设,那么构建的模型真实性就值得怀疑。 我们对常见的几种降级处理方式进行了总结。     各种方法特点总结如下: 方法 特点 推荐性 二分法 简单粗暴 馊主意 多分类 3、4或5个类别可捕获更多的预后信息,但它们不流畅,对切点的选择敏感且不稳 不推荐,解释不清楚 多项式 添加二次项、立方项 可检查非线性关系,但是不稳定 变换 对数、平方根、逆 可以提供可靠的非线性总结 分段多项式(FP) 多项式灵

版权声明:
本网站所有内容来源注明为“梅斯医学”或“MedSci原创”的文字、图片和音视频资料,版权均属于梅斯医学所有。非经授权,任何媒体、网站或个人不得转载,授权转载时须注明来源为“梅斯医学”。其它来源的文章系转载文章,或“梅斯号”自媒体发布的文章,仅系出于传递更多信息之目的,本站仅负责审核内容合规,其内容不代表本站立场,本站不负责内容的准确性和版权。如果存在侵权、或不希望被转载的媒体或个人可与我们联系,我们将立即进行删除处理。
在此留言
评论区 (5)
#插入话题
  1. [GetPortalCommentsPageByObjectIdResponse(id=1064339, encodeId=184110643394c, content=历害, beContent=null, objectType=article, channel=null, level=null, likeNumber=51, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20201203/af60709e11974ea59b6d6befa12f87e6/b0048de7bc3d4e30a52c356cca6eece2.jpg, createdBy=ed1c5440884, createdName=ms8000000746950370, createdTime=Wed Oct 27 12:10:35 CST 2021, time=2021-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1814163, encodeId=be4a18141630f, content=<a href='/topic/show?id=1a9d945490f' target=_blank style='color:#2F92EE;'>#连续变量#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=25, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=94549, encryptionId=1a9d945490f, topicName=连续变量)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=4fd6156, createdName=oliver169, createdTime=Mon Sep 27 13:55:58 CST 2021, time=2021-09-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1262807, encodeId=75bc126280e16, content=<a href='/topic/show?id=52941002359d' target=_blank style='color:#2F92EE;'>#预测模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=38, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=100235, encryptionId=52941002359d, topicName=预测模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=c3ff68, createdName=维他命, createdTime=Tue Apr 13 11:55:58 CST 2021, time=2021-04-13, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=956307, encodeId=349595630e84, content=赞赞赞, beContent=null, objectType=article, channel=null, level=null, likeNumber=73, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=b0fc5479725, createdName=酱酱125, createdTime=Mon Apr 12 20:59:07 CST 2021, time=2021-04-12, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047905, encodeId=3f69104e90561, content=梅斯里提供了很多疾病的模型计算公式,赞一个!, beContent=null, objectType=article, channel=null, level=null, likeNumber=52, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Apr 11 23:55:58 CST 2021, time=2021-04-11, status=1, ipAttribution=)]
    2021-10-27 ms8000000746950370

    历害

    0

  2. [GetPortalCommentsPageByObjectIdResponse(id=1064339, encodeId=184110643394c, content=历害, beContent=null, objectType=article, channel=null, level=null, likeNumber=51, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20201203/af60709e11974ea59b6d6befa12f87e6/b0048de7bc3d4e30a52c356cca6eece2.jpg, createdBy=ed1c5440884, createdName=ms8000000746950370, createdTime=Wed Oct 27 12:10:35 CST 2021, time=2021-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1814163, encodeId=be4a18141630f, content=<a href='/topic/show?id=1a9d945490f' target=_blank style='color:#2F92EE;'>#连续变量#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=25, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=94549, encryptionId=1a9d945490f, topicName=连续变量)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=4fd6156, createdName=oliver169, createdTime=Mon Sep 27 13:55:58 CST 2021, time=2021-09-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1262807, encodeId=75bc126280e16, content=<a href='/topic/show?id=52941002359d' target=_blank style='color:#2F92EE;'>#预测模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=38, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=100235, encryptionId=52941002359d, topicName=预测模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=c3ff68, createdName=维他命, createdTime=Tue Apr 13 11:55:58 CST 2021, time=2021-04-13, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=956307, encodeId=349595630e84, content=赞赞赞, beContent=null, objectType=article, channel=null, level=null, likeNumber=73, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=b0fc5479725, createdName=酱酱125, createdTime=Mon Apr 12 20:59:07 CST 2021, time=2021-04-12, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047905, encodeId=3f69104e90561, content=梅斯里提供了很多疾病的模型计算公式,赞一个!, beContent=null, objectType=article, channel=null, level=null, likeNumber=52, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Apr 11 23:55:58 CST 2021, time=2021-04-11, status=1, ipAttribution=)]
  3. [GetPortalCommentsPageByObjectIdResponse(id=1064339, encodeId=184110643394c, content=历害, beContent=null, objectType=article, channel=null, level=null, likeNumber=51, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20201203/af60709e11974ea59b6d6befa12f87e6/b0048de7bc3d4e30a52c356cca6eece2.jpg, createdBy=ed1c5440884, createdName=ms8000000746950370, createdTime=Wed Oct 27 12:10:35 CST 2021, time=2021-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1814163, encodeId=be4a18141630f, content=<a href='/topic/show?id=1a9d945490f' target=_blank style='color:#2F92EE;'>#连续变量#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=25, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=94549, encryptionId=1a9d945490f, topicName=连续变量)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=4fd6156, createdName=oliver169, createdTime=Mon Sep 27 13:55:58 CST 2021, time=2021-09-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1262807, encodeId=75bc126280e16, content=<a href='/topic/show?id=52941002359d' target=_blank style='color:#2F92EE;'>#预测模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=38, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=100235, encryptionId=52941002359d, topicName=预测模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=c3ff68, createdName=维他命, createdTime=Tue Apr 13 11:55:58 CST 2021, time=2021-04-13, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=956307, encodeId=349595630e84, content=赞赞赞, beContent=null, objectType=article, channel=null, level=null, likeNumber=73, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=b0fc5479725, createdName=酱酱125, createdTime=Mon Apr 12 20:59:07 CST 2021, time=2021-04-12, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047905, encodeId=3f69104e90561, content=梅斯里提供了很多疾病的模型计算公式,赞一个!, beContent=null, objectType=article, channel=null, level=null, likeNumber=52, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Apr 11 23:55:58 CST 2021, time=2021-04-11, status=1, ipAttribution=)]
  4. [GetPortalCommentsPageByObjectIdResponse(id=1064339, encodeId=184110643394c, content=历害, beContent=null, objectType=article, channel=null, level=null, likeNumber=51, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20201203/af60709e11974ea59b6d6befa12f87e6/b0048de7bc3d4e30a52c356cca6eece2.jpg, createdBy=ed1c5440884, createdName=ms8000000746950370, createdTime=Wed Oct 27 12:10:35 CST 2021, time=2021-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1814163, encodeId=be4a18141630f, content=<a href='/topic/show?id=1a9d945490f' target=_blank style='color:#2F92EE;'>#连续变量#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=25, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=94549, encryptionId=1a9d945490f, topicName=连续变量)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=4fd6156, createdName=oliver169, createdTime=Mon Sep 27 13:55:58 CST 2021, time=2021-09-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1262807, encodeId=75bc126280e16, content=<a href='/topic/show?id=52941002359d' target=_blank style='color:#2F92EE;'>#预测模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=38, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=100235, encryptionId=52941002359d, topicName=预测模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=c3ff68, createdName=维他命, createdTime=Tue Apr 13 11:55:58 CST 2021, time=2021-04-13, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=956307, encodeId=349595630e84, content=赞赞赞, beContent=null, objectType=article, channel=null, level=null, likeNumber=73, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=b0fc5479725, createdName=酱酱125, createdTime=Mon Apr 12 20:59:07 CST 2021, time=2021-04-12, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047905, encodeId=3f69104e90561, content=梅斯里提供了很多疾病的模型计算公式,赞一个!, beContent=null, objectType=article, channel=null, level=null, likeNumber=52, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Apr 11 23:55:58 CST 2021, time=2021-04-11, status=1, ipAttribution=)]
    2021-04-12 酱酱125

    赞赞赞

    0

  5. [GetPortalCommentsPageByObjectIdResponse(id=1064339, encodeId=184110643394c, content=历害, beContent=null, objectType=article, channel=null, level=null, likeNumber=51, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20201203/af60709e11974ea59b6d6befa12f87e6/b0048de7bc3d4e30a52c356cca6eece2.jpg, createdBy=ed1c5440884, createdName=ms8000000746950370, createdTime=Wed Oct 27 12:10:35 CST 2021, time=2021-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1814163, encodeId=be4a18141630f, content=<a href='/topic/show?id=1a9d945490f' target=_blank style='color:#2F92EE;'>#连续变量#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=25, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=94549, encryptionId=1a9d945490f, topicName=连续变量)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=4fd6156, createdName=oliver169, createdTime=Mon Sep 27 13:55:58 CST 2021, time=2021-09-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1262807, encodeId=75bc126280e16, content=<a href='/topic/show?id=52941002359d' target=_blank style='color:#2F92EE;'>#预测模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=38, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=100235, encryptionId=52941002359d, topicName=预测模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=c3ff68, createdName=维他命, createdTime=Tue Apr 13 11:55:58 CST 2021, time=2021-04-13, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=956307, encodeId=349595630e84, content=赞赞赞, beContent=null, objectType=article, channel=null, level=null, likeNumber=73, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=b0fc5479725, createdName=酱酱125, createdTime=Mon Apr 12 20:59:07 CST 2021, time=2021-04-12, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047905, encodeId=3f69104e90561, content=梅斯里提供了很多疾病的模型计算公式,赞一个!, beContent=null, objectType=article, channel=null, level=null, likeNumber=52, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Apr 11 23:55:58 CST 2021, time=2021-04-11, status=1, ipAttribution=)]
    2021-04-11 CHANGE

    梅斯里提供了很多疾病的模型计算公式,赞一个!

    0

相关资讯

JAHA:妊娠高血压疾病女性心血管疾病预测模型的开发和验证

该研究的模型在预测HDP女性发生CVD的10年风险方面表现良好。研究人员建议在考虑用于临床之前增加临床变量并进行外部验证。

Heart:2型糖尿病患者心血管和肾脏预后的预测模型

多种2型糖尿病患者心血管和肾脏预后的预测模型孰优孰劣?

Sci Rep:如何应对新冠二次暴发?新研究通过预测模型成功模拟欧洲二次疫情发展形势

第一次新冠疫情爆发很大程度上在于事发突然,许多防控措施无法及时实施。如今,面临二次疫情的威胁,人类是否能基于现有的数据对新冠疫情发展进行预测,从而避免或减少疫情二次冲击造成的损失呢?

Am J Otolaryngol:机器学习模型预测特发性突发感音神经性听力损失治疗后的听力恢复情况

特发性突发感音神经性听力损失(ISSHL)是一种急诊耳科疾病,其确定的预后因素仍不清楚。最近,有研究人员应用机器学习方法建立了一种新的ISSHL预后预测模型。

Hepatol Int:术前即可预测肝癌患者术后6个月内死亡率的个体化模型!

术前就能知道术后有多大的概率会死,就可以更好的决定要不要做手术了