Data reduction for X-ray serial crystallography using machine learning.
Data reduction for X-ray serial crystallography using machine learning.
复制标题
DOI：
10.1107/s1600576722011748
复制
发表时间：
2023-02-01
影响因子：
6.1
通讯作者：
Graafsma, Heinz
中科院分区：
文献类型：
--
作者：
Rahmani, Vahid;Nawaz, Shah;Pennicard, David;Setty, Shabarish Pala Ramakantha;Graafsma, Heinz
提出了一种用于区分连续结晶学图像优劣的机器学习方法。为了降低计算量，采用计算机视觉中的定向快速旋转特征提取方法来检测图像特征，然后用多层感知器(神经网络)对图像进行分类。系列结晶学实验产生了大量的实验数据。然而，尽管有这些大规模的数据集，但只有一小部分数据对下游分析有用。因此，必须可靠地区分可接受的数据(命中)和不可接受的数据(未命中)。为此，提出了一种新的数据分类流水线，该流水线从图像中提取特征，用视觉词袋方法对这些特征进行总结，然后使用机器学习对图像进行分类。此外，对各种特征提取和机器学习分类器进行了新的研究，目的是为连续的晶体数据寻找最好的特征提取和机器学习分类器。研究表明，采用多层感知器分类器的定向快速旋转简短(ORB)特征提取方法效果最好。最后，在包括合成数据和实验数据在内的各种数据集上对多层感知器的ORB特征提取方法进行了评估，与其他特征提取方法和分类器相比，该方法具有更好的性能。
A machine learning method for distinguishing good and bad images in serial crystallography is presented. To reduce the computational cost, this uses the oriented FAST and rotated BRIEF feature extraction method from computer vision to detect image features, followed by a multilayer perceptron (neural network) to classify the images. Serial crystallography experiments produce massive amounts of experimental data. Yet in spite of these large-scale data sets, only a small percentage of the data are useful for downstream analysis. Thus, it is essential to differentiate reliably between acceptable data (hits) and unacceptable data (misses). To this end, a novel pipeline is proposed to categorize the data, which extracts features from the images, summarizes these features with the ‘bag of visual words’ method and then classifies the images using machine learning. In addition, a novel study of various feature extractors and machine learning classifiers is presented, with the aim of finding the best feature extractor and machine learning classifier for serial crystallography data. The study reveals that the oriented FAST and rotated BRIEF (ORB) feature extractor with a multilayer perceptron classifier gives the best results. Finally, the ORB feature extractor with multilayer perceptron is evaluated on various data sets including both synthetic and experimental data, demonstrating superior performance compared with other feature extractors and classifiers.