Machine learning and synchrotron radiation-based micro X-ray fluorescence imaging show promise for early diagnosis of cancer by identifying trace biometals as potential cancer biomarkers. This study demonstrates the feasibility of using machine learning algorithms to analyze the spatial distribution of biometals and classify cancer development stages, providing potential advances in non-invasive cancer detection.
Although trace biometals in biological samples show potential as cancer biomarkers, spatial analysis for early cancer diagnosis remains challenging. In a recent study led by JJ Oconda and conducted at the Department of Physics, University of Nairobi, researchers utilized machine learning and synchrotron-based micro X-ray fluorescence imaging to detect trace biometals and their spatial distribution. The feasibility of locating is discussed. Distribution as a cancer biomarker.This study was published in the journal Spectrokymica Actor Part B: Atomic Spectroscopy (1).
The research team used a principal component analysis (PCA)-enabled artificial neural network (ANN) to identify biometals such as manganese (Mn), iron (Fe), copper (Cu), and selenium (Se) in the model. We determined the spatial profile at the same time. Human cell cultures (DU145 and Vero). By culturing cell lines on silicon nitride membranes and performing micro-X-ray fluorescence (micro-XRF) analysis at the TwinMic beamline, an Eletra synchrotron source, researchers obtained high-resolution 2D maps of trace biometals Did.
To analyze the complex multivariate relationships between the spatial distribution of trace biometals and the disease stages of cancer, we used PCA to reduce the data dimensionality. An ANN model trained using pixel spectral profiles of biometals successfully classified cell cultures into cancerous and healthy groups. Moreover, the selected spectral profiles allowed ANN to classify cancer cells into early, intermediate and advanced stages based on Fe and Cu fluorescence lines and Compton scattering.
In this study, we observed a high spatial correlation between Cu and Fe in cancer cell cultures compared to normal cells, demonstrating promising results. These findings pave the way for early cancer diagnosis by exploiting the spatial distribution changes and multivariate properties of trace biometals as cancer biomarkers.
The integration of machine learning and synchrotron-based micro X-ray fluorescence imaging provides a powerful tool for early detection of cancer, providing valuable insight into the complex relationship between trace biometals and stages of cancer development. It can be obtained. This research will contribute to the development of non-invasive and accurate diagnostic techniques, bringing us closer to improving cancer management and patient outcomes.
reference
(1) JJ Okonda; Anjeyo, Hong Kong. Dehayem-Camajou, A. AE Rogena The potential for early cancer diagnosis with machine learning has enabled synchrotron-based micro X-ray fluorescence imaging of trace biometals as cancer biomarkers. Spectrokymica Actor Part B: At. Spectrometer. 2023204, 106671. DOI:10.1016/j.sab.2023.106671
