Quantitative Convolutional Neural Network Based Multi-Phase XRD Pattern Analysis

X-ray diffraction (XRD) is commonly used to an-alyze phase compositions of crystalline samples. Medical ap-plications include the analysis of biotechnological materialsand gall- and kidney stones, where composition can informpathology assessment. XRD analysis methods like Rietveld re-finement requires expert knowledge, and multi-phase sampleanalysis is especially challenging and time consuming. Large-scale medical and biotechnological experiments can thereforebe hindered by the need to perform analysis tasks using XRD.Here, we present preliminary results on an automated con-volutional neural network (CNN) based method for samplecomposition analysis using XRD patterns. It can aid experts’analysis using initial estimations, and enable basic judgementsfor non-experts. Furthermore, we confirm the intuitive notionthat analysis performance degrades with sample complexitythrough systematic investigation using a synthetic dataset.