Barbu Adrian, Suehling Michael, Xu Xun, Liu David, Zhou S Kevin, Comaniciu Dorin
Statistics Department, Florida State Univ., Tallahassee, FL 32306, USA.
Med Image Comput Comput Assist Interv. 2010;13(Pt 1):28-36. doi: 10.1007/978-3-642-15705-9_4.
Lymph node detection and measurement is a difficult and important part of cancer treatment. In this paper we present a robust and effective learning-based method for the automatic detection of solid lymph nodes from Computed Tomography data. The contributions of the paper are the following. First, it presents a learning based approach to lymph node detection based on Marginal Space Learning. Second, it presents an efficient MRF-based segmentation method for solid lymph nodes. Third, it presents two new sets of features, one set self-aligning to the local gradients and another set based on the segmentation result. An extensive evaluation on 101 volumes containing 362 lymph nodes shows that this method obtains a 82.3% detection rate at 1 false positive per volume, with an average running time of 5-20 seconds per volume.
淋巴结检测与测量是癌症治疗中困难且重要的环节。在本文中，我们提出了一种基于学习的稳健且有效的方法，用于从计算机断层扫描数据中自动检测实体淋巴结。本文的贡献如下。首先，提出了一种基于边际空间学习的淋巴结检测学习方法。其次，提出了一种用于实体淋巴结的基于马尔可夫随机场的高效分割方法。第三，提出了两组新特征，一组与局部梯度自对齐，另一组基于分割结果。对包含362个淋巴结的101个容积进行的广泛评估表明，该方法在每容积1个假阳性的情况下检测率达到82.3%，每容积平均运行时间为5 - 20秒。
