Jadwal Sholat

Memuat jadwal sholatโ€ฆ

Ilmu Komputer & AI editorial

Open AccessOA2024

Metabolomic Machine Learning Predictor for Diagnosis and Prognosis of Gastric Cancer

A targeted metabolomics study of 702 plasma samples identifies a 10-metabolite diagnostic panel for gastric cancer with 0.905 sensitivity, outperforming conventional protein markers, and a prognostic model that stratifies patients into risk groups.
Yangzi Chen; Bo-Hong Wang; Yizi Zhao; Xin-Xin Shao; Ming-Shuo Wang; Fu-Hai Ma; Lai-Shou Yang; Meng Nie; P. Jin; Ke Yao; Hai-Bin Song; S. Lou; Hang Wang; Tian-Shu Yang; Yan-Tao Tian; Peng Han; Ze-Ping Huยท Nature Communicationsยท 2024ยท DOI 10.1038/s41467-024-46043-y

The core problem

Gastric cancer (GC) remains a leading cause of cancer-related mortality worldwide. Early detection is critical for improving survival, yet current methods such as endoscopy are invasive and time-consuming. Non-invasive biomarkers for early diagnosis and patient risk stratification are urgently needed. This study aims to elucidate the metabolic reprogramming in GC and develop machine learning models for diagnosis and prognosis using plasma metabolomics.

Innovation

The authors conducted a targeted metabolomics analysis of 702 plasma samples from multi-center participants. Machine learning algorithms were applied to identify metabolite signatures associated with GC. Diagnostic and prognostic models were developed and validated in external test sets. The diagnostic model was compared with conventional cancer protein markers, and the prognostic model was compared with traditional models using clinical parameters.
Introduction
Gastric cancer (GC) remains a leading cause of cancer-related mortality worldwide. Early detection is critical for improving survival, yet current methods such as endoscopy are invasive and time-consuming. Non-invasive biomarkers for early diagnosis and patient risk stratification are urgently needed. This study aims to elucidate the metabolic reprogramming in GC and develop machine learning models for diagnosis and prognosis using plasma metabolomics.

Why it matters

The findings reveal the metabolic landscape of GC and identify two distinct biomarker panels: one for early detection and one for prognosis prediction. The diagnostic panel's high sensitivity suggests it could be a valuable non-invasive tool for early GC detection, potentially reducing the need for invasive endoscopy. The prognostic model's ability to stratify patients into risk groups indicates its utility in guiding personalized treatment strategies. Collectively, these models facilitate precision medicine in GC. The study underscores the power of metabolomics combined with machine learning in oncology.

Who should read this

Pembaca Research Digest dan praktisi rumpun ini.

Opening member contentโ€ฆ