Three-dimensional approach to lung nodule detection in helical CT

被引:28
作者
Armato, SG [1 ]
Giger, ML [1 ]
Blackburn, JT [1 ]
Doi, K [1 ]
MacMahon, H [1 ]
机构
[1] Univ Chicago, Dept Radiol, Chicago, IL 60637 USA
来源
MEDICAL IMAGING 1999: IMAGE PROCESSING, PTS 1 AND 2 | 1999年 / 3661卷
关键词
computed tomography (CT); lung nodules; segmentation; three-dimensional analysis; automated classifier; feature analysis; image processing; computer-aided diagnosis (CAD);
D O I
10.1117/12.348611
中图分类号
R318 [生物医学工程];
学科分类号
0831 [生物医学工程];
摘要
We are developing an automated method for the detection of lung nodules in helical computed tomography (CT) images. This technique incorporates both two-dimensional and three-dimensional analyses to exploit the volumetric image data acquired during a CT examination. Gray-level thresholding is used to segment the lungs within the thorax. A rolling ball algorithm is applied to more accurately define the segmented lung regions. The set of segmented CT sections, which represents the complete lung volume, is iteratively thresholded, and a 10-point connectivity scheme is used to identify contiguous three-dimensional structures. Structures with volumes less than a predefined maximum value comprise the set of nodule candidates, which is then subjected to two- and three-dimensional feature analysis. To distinguish between candidates representing nodule and non-nodule structures, the values of the features are merged through linear discriminant analysis. When applied to a database of 17 helical thoracic CT cases, gray-level thresholding combined with the volume criterion detected 82% of the lung nodules. Linear discriminant analysis yielded an area under the receiver operating characteristic (ROC) curve of 0.93 in the task of distinguishing between nodule and non-nodule structures within this set of nodule candidates.
引用
收藏
页码:553 / 559
页数:7
相关论文
共 22 条
[11]
Johnson R A, 2007, Applied Multivariate Statistical Analysis: Pearson New International Edition
[12]
Kanazawa K, 1995, LECT NOTES COMPUT SC, V1024, P323
[13]
Peripheral lung cancer: Screening and detection with low-dose spiral CT versus radiography [J].
Kaneko, M ;
Eguchi, K ;
Ohmatsu, H ;
Kakinuma, R ;
Naruke, T ;
Suemasu, K ;
Moriyama, N .
RADIOLOGY, 1996, 201 (03) :798-802
[14]
Cancer statistics, 1999 [J].
Landis, SH ;
Murray, T ;
Bolden, S ;
Wingo, PA .
CA-A CANCER JOURNAL FOR CLINICIANS, 1999, 49 (01) :8-31
[15]
METZ CE, 1986, INVEST RADIOL, V21, P720, DOI 10.1097/00004424-198609000-00009
[16]
OKOMURA T, 1998, P SOC PHOTO-OPT INS, V3338, P1314
[17]
RYAN WJ, 1996, P COMP ASS RAD, P385
[18]
Mass screening for lung cancer with mobile spiral computed tomography scanner [J].
Sone, S ;
Takashima, S ;
Li, F ;
Yang, ZG ;
Honda, T ;
Maruyama, Y ;
Hasegawa, M ;
Yamanda, T ;
Kubo, K ;
Hanamura, K ;
Asakura, K .
LANCET, 1998, 351 (9111) :1242-1245
[19]
Screening for lung cancer - Another look; A different view [J].
Strauss, GM ;
Gleason, RE ;
Sugarbaker, DJ .
CHEST, 1997, 111 (03) :754-768
[20]
Computer aided diagnosis system for lung cancer based on helical CT images [J].
Toshioka, S ;
Kanazawa, K ;
Niki, N ;
Satoh, H ;
Ohmatsu, H ;
Eguchi, K ;
Moriyama, N .
IMAGE PROCESSING - MEDICAL IMAGING 1997, PTS 1 AND 2, 1997, 3034 :975-984