期刊文献+

64层螺旋CT重建层厚及重建间距与CAD肺结节检出准确性的关系分析 被引量:1

Relationship of reconstruction slice thickness and interval with accuray of computer-aided detection system:detection of pulmonary nodules by 64-slice helical CT
在线阅读 下载PDF
导出
摘要 目的:评估不同重建层厚及重建间距对Lung Care软件CAD系统检出肺结节准确性的影响.方法:纳入48例肺结节患者,每位患者重建5组数据,层厚/间距分别为:0.75/0.70mm,1/1mm,3/3mm,5/1mm,5/5mm.利用ROC曲线对比不同重建参数时CAD的检测准确性.结果:根据参考标准共检出152个肺结节.在0.75,1mm层厚时的曲线下面积(Az)值(分别为0.86,0.84)均显著高于3,5mm层厚时(3/3,5/1,5/5mm时Az值分别为0.48,0.53,0.40)(P<0.05);重叠重建(5/1mm)时,CAD检出肺结节的Az值(Az=0.53)大于5mm层距时(Az=0.40),二者无显著性差异,但明显低于1mm层厚重建(Az=0.84,P<0.05).1mm层厚重建时CAD检出肺结节的灵敏度最高(84%);0.75mm层厚重建时平均每位患者的假阳性数最低(0.56).随层厚增加,CAD检出肺结节的平均直径逐渐增大,检出小结节的数目明显下降.结论:使用薄层重建(层厚0.75mm或1mm)图像时,CAD的检出准确性显著高于厚层重建(层厚3mm或5mm)时.重叠重建(5/1mm)可以提高CAD的准确性,但其作用非常有限. AIM : To evaluate the effect of various reconstruction slice thicknesses and reconstruction intervals on diagnostic accuracy of pulmonary nodules with the computer-aided detection (CAD) system of LungCare software. METHODS: Forty-eight patients with pulmonary nodules, five sets of CT images were reconstructed in each patient by selecting five reconstruction slice thickness and interval combinations respectively: 0.75/0.70 ram, 1/1 mm,3/3 mm, 5/1 mm, 5/5 ram. The performance of the CAD system using different reconstruction combinations were compared with receiver operating characteristic curves analysis. RESULTS: 152 pulmonary nodules were detected according to the reference standard. The Az value was significant larger in 0.75 mm or 1 mm reconstruction section thickness ( Az = 0.86 and 0.84 for 0.75 mm and 1 mm, respectively) than that of 5.0 mm or 3.0 mm ( Az =0.48, 0.53 and 0.40 for 3/3 ram, 5/1 mm, 5/5 mm, respectively, P 〈 0.05 ). In the case of the overlapping reconstruction (5/1 mm) , although the Az value of CAD (Az =0.53 )was larger than that of 5 mm section interval( Az = 0.40 ) , there is no significant difference, but significantly smaller than that of 1 mm slice thickness ( P 〈 0.05 ). Sensitivity of CAD for 1 mm reconstruction section thickness(84% )was the largest among the five sets ,while the number of false-positive findings per patient for 0. 75 mm thickness (0.56) was the smallest. With the increase of slice thick- ness, the average diameter of the nodule detected by CAD was increased, but the small nodule was decreased significantly. CONCLUSION:0.75 mm or 1 mm reconstruction section thick- ness outperforms that of 3.0 mm or 5.0 mm using CAD system in detection of pulmonary nodules. Overlap reconstruction enhances the accuracy of CAD but its role is very limited.
出处 《第四军医大学学报》 CAS 北大核心 2009年第10期932-935,共4页 Journal of the Fourth Military Medical University
基金 陕西省科学技术研究发展计划(2007K12-02)
关键词 计算机辅助检测 肺结节 体层摄影术 螺旋计算机 图像重建 computer-assisted detection lung nodules tomo- graphy, spiral computed image reconstruction
  • 相关文献

参考文献9

  • 1Fraioli F,Bertoletti L,Napoli A,et al.Computer-aided detection (CAD) in lung cancer screening at chest MDCT:ROC analysis of CAD versus radiologist performance[J].J Thorac Imaging,2007,22:241-246.
  • 2Wormanns D,Fiebich M,Saidi M,et al.Automatic detection of pulmonary nodules at spiral CT:Clinical application of a computer-aided diagnosis system[J].Eur Radiol,2002,12:1052-1057.
  • 3Bae KT,Kim JS,Na YH,et al.Pulmonary nodules:Automated detection on CT images with morphologic matching algorithm-preliminary results[J].Radiology,2005,236:286-293.
  • 4Das M,Muhlenbruch G,Mahnken AH,et al.Small pulmonary nodules:effect of two computer-aided detection systems on radiologist performance[J].Radiology,2006,241:564-571.
  • 5徐岩,姜国红,马大庆.计算机辅助检测系统在X线胸片中的应用进展[J].实用放射学杂志,2007,23(6):842-845. 被引量:2
  • 6邹薇薇,刘士远.计算机辅助诊断检测系统在肺结节诊断中的应用[J].人民军医,2007,50(5):302-303. 被引量:3
  • 7Marten K,Grillhosl A,Seyfarth T,et al.Computer-assisted detection of pulmonary nodules:evaluation of diagnostic performance using an expert knowledge-based detection system with variable reconstruction slice thickness settings[J].Eur Radiol,2005,15:203-212.
  • 8Kim JS,Kim JH,Cho G,et al.Automated detection of pulmonary nodules on CT images:Effect of section thickness and reconstruction interval-initial results[J].Radiology,2005,236:295-299.
  • 9Wiemker R,Rogalla P,Zwartkruis A,et al.Computer aided lung nodule detection on high resolution CT data[J].Proceed SPIE,2002,4684:677-688.

二级参考文献26

  • 1罗清,张国华.数字化双能量减影技术在胸部肺结节及结节内钙化诊断中的应用价值[J].实用放射学杂志,2006,22(1):30-32. 被引量:20
  • 2Fiebich M,Wormanns D,Heindel W.Improvement of method for computer-assisted detection of pulmonary nodules in CT of the chest[C].Proceedings of SPIE Medical Imaging Conference,2001:702-709.
  • 3Fan L,Nowak C,Qian J,et al.Automatic detection of lung nodules from multi-slice low-dose CT images[C].Proceedings of SPIE Medical Imaging Conference,2001:1 828-1 835.
  • 4Novak C,Fan L,Qian J,et al.An interactive system for CT lung nodule identification and examination.Proceedings of International Conference on Computer Assisted Radiology and Surgery CARS 2001[C].Amsterdam,The Netherlands:Elsevier,2001:599-604.
  • 5Wiemker R,Rogalla P,Zwartkruis A,et al.Computer aided lung nodule detection on high resolution CT data[C].Proceedings of SPIE Medical Imaging,2002:677-688.
  • 6Jae L,Gordon G,Julianna C,et al.Lung nodule detection on chest CT:Evaluation of a computer-aided detection (CAD) system[J].NY Korean J Radiol,2005,6(2):89-93.
  • 7Giunta S,Ordonez PL,Canitano S,et al.Performance of a computer-aided diagnosis (CAD) system for the detection of pulmonary nodules at multidetector CT data in low dose lung cancer screening[J].Cli Imaging,2005,29:298-301.
  • 8Matthew S.Brown PhDa,Jonathan G,et al.Computer-aided lung nodule detection in CT:Results of large-scale observer testl[J].Acad Radiol,2005,12:681-686
  • 9Shah SK,McNitt-Gray MF,De Zoysa KR,et al.Solitary pulmonary nodule diagnosis on CT:Results of an observer study[J].Acad Radiol,2005,12(4):496-501.
  • 10Doi K.Overview on research and development of computer-aided diagnostic schemes[J].Semin Ultrasound CT MR,2004,25 (5):404 -410.

共引文献3

同被引文献7

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部