Automatic selection of parameters for vessel/neurite segmentation algorithms

被引:25
作者
Abdul-Karim, MA [1 ]
Roysam, B
Dowell-Mesfin, NM
Jeromin, A
Yuksel, M
Kalyanaraman, S
机构
[1] Rensselaer Polytech Inst, Troy, NY 12180 USA
[2] New York State Dept Hlth, Wadsworth Ctr, Albany, NY 12237 USA
[3] Baylor Coll Med, Houston, TX 77030 USA
基金
美国国家科学基金会; 美国国家卫生研究院;
关键词
image segmentation; minimum description length; optimization methods; segmentation evaluation;
D O I
10.1109/TIP.2005.852462
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
An automated method is presented for selecting optimal parameter settings for vessel/neurite segmentation algorithms using the minimum description length principle and a recursive random search algorithm. It trades off a probabilistic measure of image-content coverage against its conciseness. It enables nonexpert users to select parameter settings objectively, without knowledge of underlying algorithms, broadening the applicability of the segmentation algorithm, and delivering higher morphometric accuracy. It enables adaptation of parameters across batches of images. It simplifies the user interface to just one optional parameter and reduces the cost of technical support. Finally, the method is modular, extensible, and amenable to parallel computation. The method is applied to 223 images of human retinas and cultured neurons, from four different sources, using a single segmentation algorithm with eight parameters. Improvements in segmentation quality compared to default settings using 1000 iterations ranged from 4.7%-21%. Paired t-tests; showed that improvements are statistically significant (P < 0.0005). Most of the improvement occurred in the first 44 iterations. Improvements in description lengths and agreement with the ground truth were strongly correlated (p = 0.78).
引用
收藏
页码:1338 / 1350
页数:13
相关论文
共 73 条
[1]   Automated tracing and change analysis of angiogenic vasculature from in vivo multiphoton confocal image time series [J].
Abdul-Karim, MA ;
Al-Kofahi, K ;
Brown, EB ;
Jain, RK ;
Roysam, B .
MICROVASCULAR RESEARCH, 2003, 66 (02) :113-125
[2]   Median-based robust algorithms for tracing neurons from noisy confocal microscope images [J].
Al-Kofahi, KA ;
Can, A ;
Lasek, S ;
Szarowski, DH ;
Dowell-Mesfin, N ;
Shain, W ;
Turner, JT ;
Roysam, B .
IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, 2003, 7 (04) :302-317
[3]   Rapid automated three-dimensional tracing of neurons from confocal image stacks [J].
Al-Kofahi, KA ;
Lasek, S ;
Szarowski, DH ;
Pace, CJ ;
Nagy, G ;
Turner, JN ;
Roysam, B .
IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, 2002, 6 (02) :171-187
[4]   Comments on:: A methodology for evaluation of boundary detection algorithms on medical images [J].
Alberola-López, C ;
Martín-Fernández, M ;
Ruiz-Alzola, J .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2004, 23 (05) :658-660
[6]   Initialization, noise, singularities, and scale in height ridge traversal for tubular object centerline extraction [J].
Aylward, SR ;
Bullitt, E .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2002, 21 (02) :61-75
[7]   The minimum description length principle in coding and modeling [J].
Barron, A ;
Rissanen, J ;
Yu, B .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1998, 44 (06) :2743-2760
[8]  
BESAG J, 1986, J R STAT SOC B, V48, P259
[9]   Edge detector evaluation using empirical ROC curves [J].
Bowyer, K ;
Kranenburg, C ;
Dougherty, S .
COMPUTER VISION AND IMAGE UNDERSTANDING, 2001, 84 (01) :77-103
[10]  
Bühler K, 2004, MATH VISUAL, P399