Application of machine learning classifiers to X-ray diffraction imaging with medically relevant phantoms.
Application of machine learning classifiers to X-ray diffraction imaging with medically relevant phantoms.
复制标题
DOI：
10.1002/mp.15366
复制
发表时间：
2022-01
期刊：
影响因子：
3.8
通讯作者：
Greenberg JA
中科院分区：
文献类型：
--
作者：
Stryker S;Kapadia AJ;Greenberg JA
最近的研究已经证明了快速产生大视场X射线衍射图像的能力，这为理解和分析疾病提供了丰富的新数据。然而，开发算法以最大限度地提高决策和诊断任务的性能的工作才刚刚开始。在这项研究中，我们提出了基于规则和机器学习分类器的实施和比较的X射线衍射图像的医学相关的幻影，探索潜在的提高分类性能。医学相关的幻影被用来提供良好表征的地面实况比较分类器的性能。水和聚乳酸（PLA）塑料分别被用作癌组织和健康组织的替代品，并创建具有不同水平的空间复杂性和生物相关特征的模型，用于定量测试分类器性能。我们以前开发的X射线扫描仪用于获取共注册的X射线透射和衍射图像的幻影。对于分类算法，我们探索并比较了两种基于规则的分类器（互相关，或匹配滤波器，和线性最小二乘解混）和两种机器学习分类器（支持向量机和浅层神经网络）。将参考X射线衍射光谱（通过商业衍射仪测量）提供给基于规则的算法，而60%的测量的X射线衍射像素用于机器学习算法的训练。受试者工作特征曲线（AUC）下的面积用作分类算法之间的比较度量，沿着每个分类器在中点阈值处的准确度性能。材料分类的AUC值为0.994（互相关，CC）、0.994（最小二乘，LS）、0.995（支持向量机，SVM）和0.999（浅层神经网络，SNN）。将每个分类器的分类阈值设置为中点导致CC= 96.48%、LS= 96.48%、SVM= 97.36%和SNN= 98.94%的准确度值。如果仅考虑距离水-PLA边界± 3 mm的像素（由于成像分辨率限制，可能发生部分体积效应），分类准确率为CC= 89.32%，LS= 89.32%，SVM= 92.03%，SNN= 96.79%，表明机器学习算法在对成像任务至关重要的空间区域中产生了更大的改进。仅通过透射数据分类产生0.773的AUC和85.45%的准确度，远低于应用于X射线衍射图像数据的任何分类器的性能水平。我们证明了基于机器学习的分类器在整体分类准确性方面优于基于规则的方法，并提高了医学模型X射线衍射图像的空间分辨分类性能。特别是，当单个体素中存在多种材料时，机器学习算法表现出显著改善的性能。定量性能的提高展示了一种提取和利用X射线衍射成像数据的途径，以改善研究，工业和临床应用的材料分析。
Recent studies have demonstrated the ability to rapidly produce large field of view X-ray diffraction images, which provide rich new data relevant to the understanding and analysis of disease. However, work has only just begun on developing algorithms that maximize the performance towards decision-making and diagnostic tasks. In this study, we present the implementation of and comparison between rules-based and machine learning classifiers on X-ray diffraction images of medically relevant phantoms to explore the potential for increased classification performance. Medically relevant phantoms were utilized to provide well-characterized ground-truths for comparing classifier performance. Water and polylactic acid (PLA) plastic were used as surrogates for cancerous and healthy tissue, respectively, and phantoms were created with varying levels of spatial complexity and biologically relevant features for quantitative testing of classifier performance. Our previously developed X-ray scanner was used to acquire co-registered X-ray transmission and diffraction images of the phantoms. For classification algorithms, we explored and compared two rules-based classifiers (cross-correlation, or matched-filter, and linear least-squares unmixing) and two machine learning classifiers (support vector machines and shallow neural networks). Reference X-ray diffraction spectra (measured by a commercial diffractometer) were provided to the rules-based algorithms, while 60% of the measured X-ray diffraction pixels were used for training of the machine learning algorithms. The area under the receiver operating characteristic curve (AUC) was used as a comparative metric between the classification algorithms, along with the accuracy performance at the midpoint threshold for each classifier. The AUC values for material classification were 0.994 (cross-correlation, CC), 0.994 (least-squares, LS), 0.995 (support vector machine, SVM), and 0.999 (shallow neural network, SNN). Setting the classification threshold to the midpoint for each classifier resulted in accuracy values of CC=96.48%, LS=96.48%, SVM=97.36%, and SNN=98.94%. If only considering pixels ± 3 mm from water-PLA boundaries (where partial volume effects could occur due to imaging resolution limits), the classification accuracies were CC=89.32%, LS=89.32%, SVM=92.03%, and SNN=96.79%, demonstrating an even larger improvement produced by the machine-learned algorithms in spatial regions critical for imaging tasks. Classification by transmission data alone produced an AUC of 0.773 and accuracy of 85.45%, well below the performance levels of any of the classifiers applied to X-ray diffraction image data. We demonstrated that machine learning-based classifiers outperformed rules-based approaches in terms of overall classification accuracy and improved the spatially resolved classification performance on X-ray diffraction images of medical phantoms. In particular, the machine learning algorithms demonstrated considerably improved performance whenever multiple materials existed in a single voxel. The quantitative performance gains demonstrate an avenue to extract and harness X-ray diffraction imaging data to improve material analysis for research, industrial, and clinical applications.
登录
查看更多内容
影响因子：
3.8
作者：
Greenberg, Joel A.;Krishnamurthy, Kalyani;Brady, David
通讯作者：
Brady, David
影响因子：
2.4
作者：
Kern, Katie;Peerzada, Lubna;MacDonald, Carolyn
通讯作者：
MacDonald, Carolyn
影响因子：
3.8
作者：
Kiarashi, Nooshin;Nolte, Adam C.;Lo, Joseph Y.
通讯作者：
Lo, Joseph Y.
影响因子：
3.8
作者：
Hassan, Mehadi;Greenberg, Joel A.;Brady, David J.
通讯作者：
Brady, David J.
影响因子：
3.8
作者：
Batchelar, DL;Davidson, MTM;Cunningham, IA
通讯作者：
Cunningham, IA