Diagnosis Commons^™

1h-Nmr Urinary Metabolomic Profiling For Diagnosis Of Gastric Cancer, Angela W. Chan, Pascal Mercier, Daniel Schiller, Robert Bailey, Sarah Robbins, Dean T. Eurich, Michael B. Sawyer, David Broadhurst

Research outputs 2014 to 2021

Background:

Metabolomics has shown promise in gastric cancer (GC) detection. This research sought to identify whether GC has a unique urinary metabolomic profile compared with benign gastric disease (BN) and healthy (HE) patients.

Methods:

Urine from 43 GC, 40 BN, and 40 matched HE patients was analysed using ¹H nuclear magnetic resonance (¹H-NMR) spectroscopy, generating 77 reproducible metabolites (QC-RSD < 25%). Univariate and multivariate (MVA) statistics were employed. A parsimonious biomarker profile of GC vs HE was investigated using LASSO regularised logistic regression (LASSO-LR). Model performance was assessed using Receiver Operating Characteristic (ROC) curves.

Results:

GC displayed a clear discriminatory biomarker profile; the BN profile overlapped with GC and HE. LASSO-LR identified …

Prediction Models For Solitary Pulmonary Nodules Based On Curvelet Textural Features And Clinical Parameters, Jing-Jing Wang, Hai-Feng Wu, Tao Sun, Xia Li, Wei Wang, Li-Xin Tao, Da Huo, Ping-Xin Lv, Wen He, Xiu-Hua Guo

Research outputs 2013

Lung cancer, one of the leading causes of cancer-related deaths, usually appears as solitary pulmonary nodules (SPNs) which are hard to diagnose using the naked eye. In this paper, curvelet-based textural features and clinical parameters are used with three prediction models [a multilevel model, a least absolute shrinkage and selection operator (LASSO) regression method, and a support vector machine (SVM)] to improve the diagnosis of benign and malignant SPNs. Dimensionality reduction of the original curvelet-based textural features was achieved using principal component analysis. In addition, non-conditional logistical regression was used to find clinical predictors among demographic parameters and morphological features. …