X-ray absorption spectroscopy combined with machine learning for diagnosis of schistosomiasis cirrhosis

A new diagnostic technique of schistosomiasis cirrhosis based on X-ray absorption spectroscopy (XAS) was studied in this paper. Taking the liver of normal and schistosomiasis mansoni mice as samples, the incident and transmission spectra of the samples were obtained by a wide-beam X-ray spectrometry detection system based on photon counting principle, and the X-ray absorption spectra were calculated. Principal component analysis (PCA) was used to compress and visualize the normalized XAS data. The XAS processed by PCA were used as input data to train k-nearest neighbor (kNN), support vector machine (SVM) and artificial neural network (ANN). The experimental results showed that the X-ray absorption of 20-30keV energy in mouse cirrhosis was greater than that in normal mouse liver, and PCA combined with kNN, SVM or ANN can achieve a highest 10-fold cross-validation accuracy of 99.502. XAS principle combined with machine learning algorithm provides a new method for the diagnosis or stage-specific diagnosis of schistosomiasis cirrhosis (C) 2020 Elsevier Ltd. All rights reserved.