A Novel and Rapid Serum Detection Technology for Non-Invasive Screening of Gastric Cancer Based on Raman Spectroscopy Combined With Different Machine Learning Methods.

Mengya Li,Xinyu Zhan,Ke Xia,Huiyan Tian,Haiyan He,Guorong Huang,Yang Zhang,Weiling Fu,Bo Lin,Changjing Yuan

doi:10.3389/fonc.2021.665176

Abstract

Gastric cancer (GC) is the fifth most common cancer in the world and a serious threat to human health. Due to its high morbidity and mortality, a simple, rapid and accurate early screening method for GC is urgently needed. In this study, the potential of Raman spectroscopy combined with different machine learning methods was explored to distinguish serum samples from GC patients and healthy controls. Serum Raman spectra were collected from 109 patients with GC (including 35 in stage I, 14 in stage II, 35 in stage III, and 25 in stage IV) and 104 healthy volunteers matched for age, presenting for a routine physical examination. We analyzed the difference in serum metabolism between GC patients and healthy people through a comparative study of the average Raman spectra of the two groups. Four machine learning methods, one-dimensional convolutional neural network, random forest, support vector machine, and K-nearest neighbor were used to explore identifying two sets of Raman spectral data. The classification model was established by using 70% of the data as a training set and 30% as a test set. Using unseen data to test the model, the RF model yielded an accuracy of 92.8%, and the sensitivity and specificity were 94.7% and 90.8%. The performance of the RF model was further confirmed by the receiver operating characteristic (ROC) curve, with an area under the curve (AUC) of 0.9199. This exploratory work shows that serum Raman spectroscopy combined with RF has great potential in the machine-assisted classification of GC, and is expected to provide a non-destructive and convenient technology for the screening of GC patients.

Highlights

Gastric cancer (GC) is a clinically common malignant tumor of the digestive tract, accounting for 5.7% of the total new incidences of malignant tumors [1]
Since Raman spectrum signals are one-dimensional, referring to the classic model structure of LeNet-5, we developed a 1D-Convolutional Neural Network (CNN) model to identify the serum Raman spectrum of GC
The results are consistent with a previous study, which is a significant gender difference in the incidence of GC [18]

Summary

Introduction

Gastric cancer (GC) is a clinically common malignant tumor of the digestive tract, accounting for 5.7% of the total new incidences of malignant tumors [1]. Improving the diagnosis rate detection rate of early GC is of significance for reducing the mortality of GC. The clinical diagnosis of GC mainly uses CT and gastrointestinal endoscopic biopsy techniques. During CT examinations, breathing artifacts are prone to occur, affecting the diagnosis and treatment results [4]. Gastroscopic biopsy, which is the gold standard for GC diagnosis, has reliable accuracy, it is difficult to popularize it to routine screening diagnosis because gastroscopy is invasive and affected by patient compliance and operator techniques. There is an urgent need for simple and practical serum detection technology in order to help accurately identify GC patients

Objectives

Methods

Results

Discussion

Conclusion