Relationship between computer segmentation performance and computer classification performance in breast CT: A simulation study using RGI segmentation and LDA classification.
Relationship between computer segmentation performance and computer classification performance in breast CT: A simulation study using RGI segmentation and LDA classification.
复制标题
DOI：
10.1002/mp.13054
复制
发表时间：
2018-06-19
期刊：
影响因子：
3.8
通讯作者：
Boone JM
中科院分区：
文献类型：
--
作者：
Lee J;Nishikawa RM;Reiser I;Boone JM
用于乳腺癌的许多计算机辅助诊断（CADx）工具开始通过完全或半自动地分割给定的乳腺病变，然后使用从图像提取的定量特征对病变的恶性可能性进行分类。通常认为，更好的分割将导致更好的分类。然而，这一点尚未得到彻底评估。本研究的目的是评估计算机分割性能和计算机分类性能之间的关系。我们使用了85个乳腺病变（良性32个，恶性56个）从乳腺计算机断层扫描（CT）的82名妇女的情况下。我们为82个乳腺CT扫描中的每一个准备了一个平滑和一个尖锐的迭代图像重建（IIR）和一个临床重建。对于每次重建，我们通过应用15种不同的分割算法创建了15种分割结果。具体来说，我们模拟了15个分割算法，通过改变参数，在一个单一的分割算法。然后，我们通过对分割的图像结果进行定量图像特征分析，创建了15个分类结果。使用10折交叉验证，我们评估了分割和分类性能之间的关系。我们发现平滑IIR的分割和分类性能之间的正相关性较低（Pearson rho中位数= 0.18），而尖锐IIR和临床重建的两个性能之间存在中度正相关性（Pearson rho中位数= 0.4 - 0.43）。然而，我们发现在分割和分类性能的尖锐IIR和临床重建的大的变化。存在分割算法导致相似的分割性能，但相应的分类性能不同的情况。这些结果表明，分割性能的改善并不能保证相应的分类性能的改善。计算机分割是影响计算机分类的间接变量。由于更好的分割并不能保证更好的分类，因此在比较分割算法时，我们应该报告分割和分类性能。
Many computer aided diagnosis (CADx) tools for breast cancer begin by fully or semi automatically segmenting a given breast lesion and then classifying the lesion’s likelihood of malignancy using quantitative features extracted from the image. It is often assumed that better segmentation will result in better classification. However, this has not been thoroughly evaluated. The purpose of this study is to evaluate the relationship between computer segmentation performance and computer classification performance. We used 85 breast lesions (32 benign, 56 malignant) from breast computed tomography (CT) cases of 82 women. We prepared one smooth and one sharp iterative image reconstructions (IIR) and a clinical reconstruction for each of the 82 breast CT scans. For each reconstruction, we created 15 segmentation outcomes by applying 15 different segmentation algorithms. Specifically, we simulated 15 segmentation algorithms by changing parameters in a single segmentation algorithm. We then created 15 classification outcomes by conducting quantitative image feature analysis on the segmented image results. Using a 10 fold cross-validation, we evaluated the relationship between segmentation and classification performances. We found a low positive correlation between segmentation and classification performances for the smooth IIR (median Pearson’s rho = 0.18), while a moderate positive correlation (median Pearson’s rho = 0.4 – 0.43) was found between the two performances for the sharp IIR and clinical reconstruction. However, we found large variations in both segmentation and classification performances for the sharp IIR and clinical reconstruction. There were cases where segmentation algorithms resulted in similar segmentation performances, but the corresponding classification performances were different. These results indicate that an improvement in segmentation performance does not guarantee an improvement in the corresponding classification performance. Computer segmentation is an indirect variable affecting the computer classification. As better segmentation does not guarantee better classification, we should report both segmentation and classification performances when comparing segmentation algorithms.
登录
查看更多内容
影响因子：
4.8
作者：
DICE, LR
通讯作者：
DICE, LR
影响因子：
3.8
作者：
Lee, Juhun;Nishikawa, Robert M.;Lindfors, Karen K.
通讯作者：
Lindfors, Karen K.
影响因子：
7.6
作者：
Marcomini, Karem D.;Carneiro, Antonio A. O.;Schiabel, Homero
通讯作者：
Schiabel, Homero
影响因子：
10.9
作者：
Rahmati, Peyman;Adler, Andy;Hamarneh, Ghassan
通讯作者：
Hamarneh, Ghassan
影响因子：
3.8
作者：
Tao, Yimo;Lo, Shih-Chung B.;Xuan, Jianhua
通讯作者：
Xuan, Jianhua