乳腺X线图像肿块分割

Segmentation of masses in mammograms

乳腺肿块分割是乳腺癌计算机辅助诊断(CAD)检测和识别系统中关键的一步。由于乳腺肿块与背景相互交叠、边界不清晰、乳房密度不均匀,使得其分割比较困难。本文基于区域增长算法,研究了利用乳腺肿块自身特征得到最优分割阈值的方法,从而提出一种对乳腺X线图像肿块快速、有效的分割方法。实验结果表明该方法在保证肿块针状化特征情况下,拥有较好的分割效果。

Segmentation is the vital step in computer-aided diagnosis on masses of mammograms. A challenge for mass segmentation in mammograms is that masses may connect with some surrounding tissues which have the similar intensity,and the intensities in mammograms are not symmetrical. In this paper, a novel improved region growing-based algorithm is proposed. In this algorithm, some important features of masses are utilized to get the best segmentation threshold. The experiments show good results with keeping the spiculation of masses, which is a primary sign of malignancy for masses.