Download This Paper
Preprints with The Lancet is part of SSRN´s First Look, a place where journals identify content of interest prior to publication. Authors have opted in at submission to The Lancet family of journals to post their preprints on Preprints with The Lancet. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early stage research papers that have not been peer-reviewed. The findings should not be used for clinical or public health decision making and should not be presented to a lay audience without highlighting that they are preliminary and have not been peer-reviewed. For more information on this collaboration, see the comments published in The Lancet about the trial period, and our decision to make this a permanent offering, or visit The Lancet´s FAQ page, and for any feedback please contact preprints@lancet.com.
Machine Learning to Serum Metabolic Fingerprints for Tuberculosis and Drug-Resistant Tuberculosis Diagnosis: An Observational Study
30 Pages Posted: 24 Feb 2023
More...Abstract
Background: Tuberculosis (TB) remains a leading cause of mortality due to a single infectious agent worldwide. Metabolic profiling can be used to detect the end products of complex metabolic reactions. However, available metabolic diagnostics for accurate TB and drug-resistant TB screening is limited. We aimed to provide efficient metabolic tools for TB screening, diagnosis, and drug resistance tests.
Methods: In this prospective observational study, patients with pulmonary TB were recruited from two hospitals in China. Nanoparticle enhanced laser desorption/ionization mass spectrometry (NPELDI MS) based on machine learning (ML) was used to develop the serum metabolic fingerprints of TB and rifampicin-resistant TB (RR-TB) patients.
Findings: Between Sep 9, 2020 and Mar 9, 2021, 228 patients were recruited, including 118 healthy subjects and 110 patients with pulmonary TB. Among them, 30 patients with RR-TB and 28 patients with RS-TB were recruited. We obtained an AUC of up to 0.978 in the differentiation of TB patients from controls, and an AUC of up to 0·857 in the differentiation of RR-TB patients from rifampicin-sensitive TB patients using ML algorithms based on biomarkers.
Interpretation: This study unveiled serum metabolic biomarker panels of TB and RR-TB for the first time by integrating ML and NPELDI MS. New diagnostic models based on biomarkers were constructed to characterize different types of TB with desirable diagnostic performance. Our work provides efficient diagnostic tools for TB and drug-resistant TB.
Trial Registration: This study was registered at ClinicalTrials.gov (NCT04490746).
Funding: This study was supported by the National Natural Science Foundation of China (Grant Number 82001818, 82230069, and 82030048), Key Research and Development Program of Zhejiang Province (Grant Number 2019C03077), Zhejiang Science and Technology Project (LQ21H180007).
Declaration of Interest: The authors declare no competing interests.
Ethical Approval: The study was carried out according to the declaration of Helsinki (1975) and approved by the Ethics Committee of Zhejiang University School of Medicine, China. Written informed consent from each participant was obtained in this study.
Keywords: Tuberculosis, drug resistant tuberculosis, metabolic fingerprinting, nanoparticle enhanced laser desorption/ionization (NPELDI) mass spectrometry, machine learning
Suggested Citation: Suggested Citation