AUTHOR=Hu Xiefei , Zhi Shenshen , Wu Wenyan , Tao Yang , Zhang Yuanyuan , Li Lijuan , Li Xun , Pan Liyan , Fan Haiping , Li Wei 

TITLE=The application of metagenomics, radiomics and machine learning for diagnosis of sepsis

JOURNAL=Frontiers in Medicine

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2024.1400166

DOI=10.3389/fmed.2024.1400166

ISSN=2296-858X

ABSTRACT=Sepsis seriously threatens the life and health of individuals. Early and accessible diagnosis and targeted treatment of this condition are of vital importance. This study aimed to sequence blood samples from sepsis patients to explore the relationship between microbes, metabolic pathways, and relevant blood test indicators. Additionally, machine learning algorithms were employed to develop a model based on medical records and radiomic features to aid in the clinical diagnosis of sepsis. The results of α -diversity and β -diversity analyses showed that the microbial diversity of sepsis group was significantly higher than that of normal group (p < 0.05). The top 10 microbial abundances in the sepsis and normal groups were Vitis vinifera, Mycobacterium canettii, Solanum pennellii, Ralstonia insidiosa, Ananas comosus, Moraxella osloensis, Escherichia coli, Staphylococcus hominis, Camelina sativa, and Cutibacterium acnes. The enriched metabolic pathways mainly included Protein families: genetic information processing, Translation, Protein families: signaling and cellular processes, and Unclassified: genetic information processing. The correlation analysis revealed a significant positive correlation (p < 0.05) between IL-6 and Membrane transport. Metabolism of other amino acids showed a significant positive correlation (p < 0.05) with Cutibacterium acnes, Ralstonia insidiosa, Moraxella osloensis, and Staphylococcus hominis. Ananas comosus showed a significant positive correlation (p < 0.05) with Poorly characterized and Unclassified: metabolism. Blood test-related indicators showed a significant negative correlation (p < 0.05) with microorganisms. Logistic regression (LR) was used as the optimal model in six machine learning models based on medical records and radiomic features. The nomogram, calibration curves, and AUC values demonstrated that LR performed best for prediction.