European Radiology:基于x线组学的骨良恶肿瘤的机器学习模型

2022-08-13 shaosai MedSci原创

放射线组学利用对多种成像特征的提取以描述肿瘤的特征[,可以作为机器学习模型的输入对肿瘤进行分类。

传统的平片检查被认为是诊断骨肿瘤和肿瘤样病变的首选成像方式,可以准确地观察到骨质破坏和骨膜反应,因此传统的平片对骨肿瘤和肿瘤样病变的诊断工作和后续治疗至关重要。磁共振成像可以通过显示骨外组织成分或肿瘤的组成来帮助缩小鉴别范围。然而,平片依然是是的首选成像方法,并具有高空间分辨率、成本效益和可性。

为了使平片上的骨病变评估标准化,可以使用计算机辅助提取成像特征的方法来进行。放射线组学利用对多种成像特征的提取以描述肿瘤的特征[,可以作为机器学习模型的输入对肿瘤进行分类。机器学习模型包括统计模型、决策树模型、支持向量机和人工神经网络(ANN)。决策树如随机森林分类器(RFC)或统计方法如逻辑回归或高斯奈夫贝叶斯分类器(GNB)被广泛用于分类任务。

近日,发表在European Radiology杂志开发并验证了一个使用从平片和人口信息中得出的放射组学特征的以区分良性和恶性骨病变机器学习模型,并在外部测试集上与放射科医生进行了性能比较,为临床快速、准确的对骨病变进行定性诊断提供了技术支持。

本研究对880名被诊断为恶性(n = 213,24.2%)或良性(n = 667,75.8%)原发性骨肿瘤的患者(年龄33.1±19.4岁,395名女性)进行了术前平片检查,并通过组织病理学确定诊断。数据被分成70%/15%/15%用于训练、验证和内部测试。此外,还从另一个机构获得了96名患者用于外部测试。使用放射学特征和人口统计学信息开发并验证了机器学习模型。每个模型的性能都在测试集上进行了评估,包括准确性、受试者工作特性的曲线下面积(AUC)、敏感性和特异性。为了比较,外部测试集由两名放射科住院医生和两名专门从事肌肉骨骼肿瘤成像的放射科医生进行评估。 

最好的机器学习模型是基于人工神经网络(ANN),结合放射学和人口学信息,在内部和外部测试集上分别达到80%和75%的准确性,75%和90%的敏感性,AUC为0.79和0.90。相比之下,放射科住院医师在61%和35%的灵敏度下达到了71%和65%的准确率,而专门从事肌肉骨骼肿瘤成像的放射科医师在90%和81%的灵敏度下分别达到了84%和83%的准确率。  


 A和B是来自一个33岁男性的胫骨恶性肿瘤的示例软骨肉瘤。A显示的是X线片,B显示的是为提取放射线组学而进行的分割。结合人口学和放射学信息的人工神经网络模型正确预测了恶性肿瘤,其确定性为86%。C和D 来自一个15岁男性的胫骨近端良性肿瘤的示例骨化性纤维瘤。A显示的是X线片,B显示的是用于放射组学提取的分割。人工神经网络模型使用人口统计学和放射学信息的组合,正确预测了一个良性肿瘤,确定性为93%

本研究表明,利用放射学特征和人口学信息建立的机器学习模型在平片上鉴别良性和恶性骨肿瘤方面显示出较高的准确性和判别能力,可以提高临床的诊断决策并改善骨肿瘤的诊断工作。

原文出处:

Claudio E von Schacky,Nikolas J Wilhelm,Valerie S Schäfer,et al.Development and evaluation of machine learning models based on X-ray radiomics for the classification and differentiation of malignant and benign bone tumors.DOI:10.1007/s00330-022-08764-w

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=1688365, encodeId=171d1688365cb, content=<a href='/topic/show?id=fab46121efa' target=_blank style='color:#2F92EE;'>#机器学习模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=29, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61217, encryptionId=fab46121efa, topicName=机器学习模型)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=2bea28558433, createdName=ms2694732865965676, createdTime=Mon Nov 14 09:04:51 CST 2022, time=2022-11-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1729722, encodeId=97fc1e2972299, content=<a href='/topic/show?id=986913980dd' target=_blank style='color:#2F92EE;'>#PE#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=22, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=13980, encryptionId=986913980dd, topicName=PE)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f06c33759375, createdName=feather89, createdTime=Thu Apr 06 09:04:51 CST 2023, time=2023-04-06, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1241519, encodeId=257b1241519b2, content=<a href='/topic/show?id=7d806121594' target=_blank style='color:#2F92EE;'>#机器学习#</a><a href='/topic/show?id=c30010183810' target=_blank style='color:#2F92EE;'>#骨肿瘤#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=49, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61215, encryptionId=7d806121594, topicName=机器学习), TopicDto(id=101838, encryptionId=c30010183810, topicName=骨肿瘤)], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20220519/c2ab253484ee4527a2d4e9589a4821ac/45de9bf494a54becb2ea4369c9d11e85.jpg, createdBy=7a3710, createdName=循证小兵, createdTime=Sun Aug 28 14:46:33 CST 2022, time=2022-08-28, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1298075, encodeId=436d12980e517, content=<a href='/topic/show?id=c42d18892ad' target=_blank style='color:#2F92EE;'>#X线#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=33, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=18892, encryptionId=c42d18892ad, topicName=X线)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=5c64275, createdName=piaojinhua, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1583378, encodeId=449c15833e8a9, content=<a href='/topic/show?id=f6e1611840d' target=_blank style='color:#2F92EE;'>#机器#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=43, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61184, encryptionId=f6e1611840d, topicName=机器)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=13dc16973127, createdName=ms6279672939590805, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=)]
  2. [GetPortalCommentsPageByObjectIdResponse(id=1688365, encodeId=171d1688365cb, content=<a href='/topic/show?id=fab46121efa' target=_blank style='color:#2F92EE;'>#机器学习模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=29, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61217, encryptionId=fab46121efa, topicName=机器学习模型)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=2bea28558433, createdName=ms2694732865965676, createdTime=Mon Nov 14 09:04:51 CST 2022, time=2022-11-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1729722, encodeId=97fc1e2972299, content=<a href='/topic/show?id=986913980dd' target=_blank style='color:#2F92EE;'>#PE#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=22, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=13980, encryptionId=986913980dd, topicName=PE)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f06c33759375, createdName=feather89, createdTime=Thu Apr 06 09:04:51 CST 2023, time=2023-04-06, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1241519, encodeId=257b1241519b2, content=<a href='/topic/show?id=7d806121594' target=_blank style='color:#2F92EE;'>#机器学习#</a><a href='/topic/show?id=c30010183810' target=_blank style='color:#2F92EE;'>#骨肿瘤#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=49, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61215, encryptionId=7d806121594, topicName=机器学习), TopicDto(id=101838, encryptionId=c30010183810, topicName=骨肿瘤)], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20220519/c2ab253484ee4527a2d4e9589a4821ac/45de9bf494a54becb2ea4369c9d11e85.jpg, createdBy=7a3710, createdName=循证小兵, createdTime=Sun Aug 28 14:46:33 CST 2022, time=2022-08-28, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1298075, encodeId=436d12980e517, content=<a href='/topic/show?id=c42d18892ad' target=_blank style='color:#2F92EE;'>#X线#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=33, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=18892, encryptionId=c42d18892ad, topicName=X线)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=5c64275, createdName=piaojinhua, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1583378, encodeId=449c15833e8a9, content=<a href='/topic/show?id=f6e1611840d' target=_blank style='color:#2F92EE;'>#机器#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=43, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61184, encryptionId=f6e1611840d, topicName=机器)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=13dc16973127, createdName=ms6279672939590805, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=)]
    2023-04-06 feather89
  3. [GetPortalCommentsPageByObjectIdResponse(id=1688365, encodeId=171d1688365cb, content=<a href='/topic/show?id=fab46121efa' target=_blank style='color:#2F92EE;'>#机器学习模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=29, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61217, encryptionId=fab46121efa, topicName=机器学习模型)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=2bea28558433, createdName=ms2694732865965676, createdTime=Mon Nov 14 09:04:51 CST 2022, time=2022-11-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1729722, encodeId=97fc1e2972299, content=<a href='/topic/show?id=986913980dd' target=_blank style='color:#2F92EE;'>#PE#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=22, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=13980, encryptionId=986913980dd, topicName=PE)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f06c33759375, createdName=feather89, createdTime=Thu Apr 06 09:04:51 CST 2023, time=2023-04-06, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1241519, encodeId=257b1241519b2, content=<a href='/topic/show?id=7d806121594' target=_blank style='color:#2F92EE;'>#机器学习#</a><a href='/topic/show?id=c30010183810' target=_blank style='color:#2F92EE;'>#骨肿瘤#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=49, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61215, encryptionId=7d806121594, topicName=机器学习), TopicDto(id=101838, encryptionId=c30010183810, topicName=骨肿瘤)], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20220519/c2ab253484ee4527a2d4e9589a4821ac/45de9bf494a54becb2ea4369c9d11e85.jpg, createdBy=7a3710, createdName=循证小兵, createdTime=Sun Aug 28 14:46:33 CST 2022, time=2022-08-28, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1298075, encodeId=436d12980e517, content=<a href='/topic/show?id=c42d18892ad' target=_blank style='color:#2F92EE;'>#X线#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=33, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=18892, encryptionId=c42d18892ad, topicName=X线)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=5c64275, createdName=piaojinhua, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1583378, encodeId=449c15833e8a9, content=<a href='/topic/show?id=f6e1611840d' target=_blank style='color:#2F92EE;'>#机器#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=43, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61184, encryptionId=f6e1611840d, topicName=机器)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=13dc16973127, createdName=ms6279672939590805, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=)]
  4. [GetPortalCommentsPageByObjectIdResponse(id=1688365, encodeId=171d1688365cb, content=<a href='/topic/show?id=fab46121efa' target=_blank style='color:#2F92EE;'>#机器学习模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=29, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61217, encryptionId=fab46121efa, topicName=机器学习模型)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=2bea28558433, createdName=ms2694732865965676, createdTime=Mon Nov 14 09:04:51 CST 2022, time=2022-11-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1729722, encodeId=97fc1e2972299, content=<a href='/topic/show?id=986913980dd' target=_blank style='color:#2F92EE;'>#PE#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=22, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=13980, encryptionId=986913980dd, topicName=PE)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f06c33759375, createdName=feather89, createdTime=Thu Apr 06 09:04:51 CST 2023, time=2023-04-06, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1241519, encodeId=257b1241519b2, content=<a href='/topic/show?id=7d806121594' target=_blank style='color:#2F92EE;'>#机器学习#</a><a href='/topic/show?id=c30010183810' target=_blank style='color:#2F92EE;'>#骨肿瘤#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=49, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61215, encryptionId=7d806121594, topicName=机器学习), TopicDto(id=101838, encryptionId=c30010183810, topicName=骨肿瘤)], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20220519/c2ab253484ee4527a2d4e9589a4821ac/45de9bf494a54becb2ea4369c9d11e85.jpg, createdBy=7a3710, createdName=循证小兵, createdTime=Sun Aug 28 14:46:33 CST 2022, time=2022-08-28, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1298075, encodeId=436d12980e517, content=<a href='/topic/show?id=c42d18892ad' target=_blank style='color:#2F92EE;'>#X线#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=33, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=18892, encryptionId=c42d18892ad, topicName=X线)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=5c64275, createdName=piaojinhua, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1583378, encodeId=449c15833e8a9, content=<a href='/topic/show?id=f6e1611840d' target=_blank style='color:#2F92EE;'>#机器#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=43, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61184, encryptionId=f6e1611840d, topicName=机器)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=13dc16973127, createdName=ms6279672939590805, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=)]
    2022-08-14 piaojinhua
  5. [GetPortalCommentsPageByObjectIdResponse(id=1688365, encodeId=171d1688365cb, content=<a href='/topic/show?id=fab46121efa' target=_blank style='color:#2F92EE;'>#机器学习模型#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=29, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61217, encryptionId=fab46121efa, topicName=机器学习模型)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=2bea28558433, createdName=ms2694732865965676, createdTime=Mon Nov 14 09:04:51 CST 2022, time=2022-11-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1729722, encodeId=97fc1e2972299, content=<a href='/topic/show?id=986913980dd' target=_blank style='color:#2F92EE;'>#PE#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=22, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=13980, encryptionId=986913980dd, topicName=PE)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f06c33759375, createdName=feather89, createdTime=Thu Apr 06 09:04:51 CST 2023, time=2023-04-06, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1241519, encodeId=257b1241519b2, content=<a href='/topic/show?id=7d806121594' target=_blank style='color:#2F92EE;'>#机器学习#</a><a href='/topic/show?id=c30010183810' target=_blank style='color:#2F92EE;'>#骨肿瘤#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=49, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61215, encryptionId=7d806121594, topicName=机器学习), TopicDto(id=101838, encryptionId=c30010183810, topicName=骨肿瘤)], attachment=null, authenticateStatus=null, createdAvatar=https://img.medsci.cn/20220519/c2ab253484ee4527a2d4e9589a4821ac/45de9bf494a54becb2ea4369c9d11e85.jpg, createdBy=7a3710, createdName=循证小兵, createdTime=Sun Aug 28 14:46:33 CST 2022, time=2022-08-28, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1298075, encodeId=436d12980e517, content=<a href='/topic/show?id=c42d18892ad' target=_blank style='color:#2F92EE;'>#X线#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=33, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=18892, encryptionId=c42d18892ad, topicName=X线)], attachment=null, authenticateStatus=null, createdAvatar=, createdBy=5c64275, createdName=piaojinhua, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=1583378, encodeId=449c15833e8a9, content=<a href='/topic/show?id=f6e1611840d' target=_blank style='color:#2F92EE;'>#机器#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=43, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=61184, encryptionId=f6e1611840d, topicName=机器)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=13dc16973127, createdName=ms6279672939590805, createdTime=Sun Aug 14 09:04:51 CST 2022, time=2022-08-14, status=1, ipAttribution=)]

相关资讯

European Radiology:放射组学实现GGO肺腺癌的EGFR突变状态的无创预测!

先进的图像分析算法使得通过提取大量的图像特征来自动量化成像表型成为可能,并且可以提供比眼睛辨别的表型更详细的特征,被称为 "放射组学"。

European Radiology:如何运用CT评估膀胱癌的肌肉侵犯程度?

组织病理学是诊断BCa肌肉侵犯的金标准。然而,活组织检查依赖于操作者,而且不可能对肿瘤的每个部分都进行取样。

Euro Radiol:来自不同机构、不同扫描仪的CT放射组学特征是否具有可重复性?

双能量CT(DECT)具有第二个X射线光谱,可以区分多种材料,可生成一组虚拟单色图像(VMIs),使得一些新的和临床相关的CT应用成为可能。

European Radiology:放射组学在胃癌精准医疗中的机遇与挑战

近十年来,放射组学已经成为一个备受关注的研究领域。放射组学是一个优化现有成像资源的新兴领域,其可从医学图像中高通量提取定量特征,通过预定的算法进一步分析,为临床决策支持开发模型。

European Radiology:在低级别胶质瘤中识别IDH突变和ATRX表达缺失的放射组学列线图

MRI作为一种非侵入性影像学技术,被常规用于诊断和绘制胶质瘤。有研究表明,MRI和放射组学的结合可以预测胶质瘤的等级和分子亚型。

European Radiology:图像采集和处理对超声放射组学特征再现性的影响

放射组学特征的可重复性分析可以通过评估成像数据可重复性、分割可重复性、计算/统计可重复性和研究可重复性来确定。