Thorax:使用NLST数据对Brock模型进行外部验证和重新校准,以预测肺结节中癌症的概率

2019-03-24 xiangting MedSci原创

虽然Brock模型在NLST数据集上验证时达到了高AUC,但该模型得益于更新和重新校准。

这项研究使用全国肺部筛查试验(NLST)数据集对Brock模型进行了外部验证,遵循针对个体预后或诊断多因素预测模型透明报告的严格指导。这篇文章报道了如何解释外部验证结果,并突出了重新校准和模型更新的作用。

研究人员使用NLST数据集评估模型的辨别度和校准。根据McWilliams 等人报告的纳入/排除标准,确定了基线低剂量CT筛查发现的7879例非钙化结节,并随访2年。研究描述了泛加拿大肺癌早期检测研究与NLST队列之间的差异。通过将原始Brock模型拟合到NLST来计算预后指数的斜率和截距系数。研究还评估了模型重新校准和添加新协变量的影响,如体重指数、吸烟状况、包年和石棉。

虽然该模型的曲线下面积(AUC)良好,为0.90595CI 0.8820.928),但直方图显示该模型区别良性和恶性病例的能力差。校准曲线显示该模型高估了癌症的可能性。在重新校准模型中,更新了肺气肿,毛刺征和结节计数的系数。更新模型的校准力得到改善并且乐观校正的AUC0.91295CI 0.8910.932)。在评估的新协变量中,仅发现吸烟史有显著性(p<0.01)。

虽然Brock模型在NLST数据集上验证时达到了高AUC,但该模型得益于更新和重新校准。然而,模型中使用的协变量不足以充分辨别恶性病例。

原始出处:

Audrey Winter. External validation and recalibration of the Brock model to predict probability of cancer in pulmonary nodules using NLST data. Thorax. 21 March 2019.

本文系梅斯医学(MedSci)原创编译整理,转载需授权!


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 (4)
#插入话题
  1. [GetPortalCommentsPageByObjectIdResponse(id=1710527, encodeId=29001e10527c8, content=<a href='/topic/show?id=750115604e7' target=_blank style='color:#2F92EE;'>#ROCK#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=30, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15604, encryptionId=750115604e7, topicName=ROCK)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=a19831855420, createdName=12498dabm87暂无昵称, createdTime=Sun Oct 27 12:23:00 CST 2019, time=2019-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1976993, encodeId=a31219e6993a7, content=<a href='/topic/show?id=b9f11560065' target=_blank style='color:#2F92EE;'>#ROC#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=24, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15600, encryptionId=b9f11560065, topicName=ROC)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=d590125, createdName=linlin2312, createdTime=Wed Apr 24 15:23:00 CST 2019, time=2019-04-24, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1981828, encodeId=d7b2198182830, content=<a href='/topic/show?id=76293e436d' target=_blank style='color:#2F92EE;'>#Brock模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=21, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=3743, encryptionId=76293e436d, topicName=Brock模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=8409354, createdName=闆锋旦, createdTime=Tue Oct 29 15:23:00 CST 2019, time=2019-10-29, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047686, encodeId=20a6104e68631, 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=49, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Mar 24 18:23:00 CST 2019, time=2019-03-24, status=1, ipAttribution=)]
  2. [GetPortalCommentsPageByObjectIdResponse(id=1710527, encodeId=29001e10527c8, content=<a href='/topic/show?id=750115604e7' target=_blank style='color:#2F92EE;'>#ROCK#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=30, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15604, encryptionId=750115604e7, topicName=ROCK)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=a19831855420, createdName=12498dabm87暂无昵称, createdTime=Sun Oct 27 12:23:00 CST 2019, time=2019-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1976993, encodeId=a31219e6993a7, content=<a href='/topic/show?id=b9f11560065' target=_blank style='color:#2F92EE;'>#ROC#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=24, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15600, encryptionId=b9f11560065, topicName=ROC)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=d590125, createdName=linlin2312, createdTime=Wed Apr 24 15:23:00 CST 2019, time=2019-04-24, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1981828, encodeId=d7b2198182830, content=<a href='/topic/show?id=76293e436d' target=_blank style='color:#2F92EE;'>#Brock模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=21, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=3743, encryptionId=76293e436d, topicName=Brock模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=8409354, createdName=闆锋旦, createdTime=Tue Oct 29 15:23:00 CST 2019, time=2019-10-29, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047686, encodeId=20a6104e68631, 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=49, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Mar 24 18:23:00 CST 2019, time=2019-03-24, status=1, ipAttribution=)]
    2019-04-24 linlin2312
  3. [GetPortalCommentsPageByObjectIdResponse(id=1710527, encodeId=29001e10527c8, content=<a href='/topic/show?id=750115604e7' target=_blank style='color:#2F92EE;'>#ROCK#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=30, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15604, encryptionId=750115604e7, topicName=ROCK)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=a19831855420, createdName=12498dabm87暂无昵称, createdTime=Sun Oct 27 12:23:00 CST 2019, time=2019-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1976993, encodeId=a31219e6993a7, content=<a href='/topic/show?id=b9f11560065' target=_blank style='color:#2F92EE;'>#ROC#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=24, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15600, encryptionId=b9f11560065, topicName=ROC)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=d590125, createdName=linlin2312, createdTime=Wed Apr 24 15:23:00 CST 2019, time=2019-04-24, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1981828, encodeId=d7b2198182830, content=<a href='/topic/show?id=76293e436d' target=_blank style='color:#2F92EE;'>#Brock模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=21, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=3743, encryptionId=76293e436d, topicName=Brock模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=8409354, createdName=闆锋旦, createdTime=Tue Oct 29 15:23:00 CST 2019, time=2019-10-29, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047686, encodeId=20a6104e68631, 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=49, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Mar 24 18:23:00 CST 2019, time=2019-03-24, status=1, ipAttribution=)]
  4. [GetPortalCommentsPageByObjectIdResponse(id=1710527, encodeId=29001e10527c8, content=<a href='/topic/show?id=750115604e7' target=_blank style='color:#2F92EE;'>#ROCK#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=30, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15604, encryptionId=750115604e7, topicName=ROCK)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=a19831855420, createdName=12498dabm87暂无昵称, createdTime=Sun Oct 27 12:23:00 CST 2019, time=2019-10-27, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1976993, encodeId=a31219e6993a7, content=<a href='/topic/show?id=b9f11560065' target=_blank style='color:#2F92EE;'>#ROC#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=24, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=15600, encryptionId=b9f11560065, topicName=ROC)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=d590125, createdName=linlin2312, createdTime=Wed Apr 24 15:23:00 CST 2019, time=2019-04-24, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1981828, encodeId=d7b2198182830, content=<a href='/topic/show?id=76293e436d' target=_blank style='color:#2F92EE;'>#Brock模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=21, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=3743, encryptionId=76293e436d, topicName=Brock模型)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=8409354, createdName=闆锋旦, createdTime=Tue Oct 29 15:23:00 CST 2019, time=2019-10-29, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1047686, encodeId=20a6104e68631, 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=49, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f0620, createdName=CHANGE, createdTime=Sun Mar 24 18:23:00 CST 2019, time=2019-03-24, status=1, ipAttribution=)]
    2019-03-24 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

这名患者中,肺癌引起了什么并发症?

患者,男,70岁,进行性头、颈、手臂和上胸部水肿伴呼吸困难、颈胸静脉扩张而入院。患者行18F-FDG PET/CT显像。

Radiology:长期低剂量CT在肺癌筛查阴性后随访的必要性

本研究旨在评价肺癌筛查阴性后患者的肺癌发生率。

哈气就能检测肺癌?上海这个实验室将于3个月后开展临床试验

“哈气”就能检测出肺癌吗?2月18日,国内首个呼吸活检实验室落户上海交通大学医学院附属仁济医院,实验室由仁济医院与英国相关机构合作开设,将于3个月后正式开展肺癌早期检测的临床试验工作。与尿素“呼吸试验”原理完全不同仁济医院肿瘤科主任、实验室负责人王理伟坦言, 近来年,我国肺癌的发病率逐年在升高,并且成为死亡率居首的恶性肿瘤。呼吸活检可以作为早期发现呼吸道癌症疾的无创检测的关键手段,不仅能降低患者的

Thorax:肺癌后第二原发癌和更高级吸烟相关原发癌的发病率

从初次诊断开始,肺癌幸存者后续肺癌、喉癌、头颈癌和食管鳞状细胞癌发病率升高至少持续10年。

JAMA Surg:退伍军人事务部肺癌视频辅助胸腔镜肺切除术的使用

与教学医院一样,退伍军人事务部中大部分的肺切除术采用视频辅助胸腔镜手术完成。

Thorax:影像学肺气肿在肺癌风险评估中的作用

影像学肺气肿是肺癌诊断的独立预测因子,其有助于指导合适患者进一步筛查的决策。