Helicobacter Pylori Infection as the Predominant High-Risk Factor for Gastric Cancer Recurrence Post-Gastrectomy: An...

Area of Science:

• Oncology
• Bioinformatics
• Medical Machine Learning

Background:

• Gastric cancer recurrence is a significant postoperative complication impacting patient outcomes.
• Identifying high-risk factors is crucial for effective patient management and personalized treatment.

Purpose of the Study:

• To develop a machine learning algorithm for identifying high-risk factors of gastric cancer recurrence.
• To investigate the correlation between gastric cancer recurrence and Helicobacter pylori (H. pylori) infection.

Main Methods:

• Utilized data from 1234 gastric cancer patients with 37 variables.
• Implemented and compared four machine learning algorithms: XGBoost, RF, KNN, and MLP.
• Validated models using k-fold cross-validation and independent datasets.

Main Results:

• XGBoost demonstrated superior predictive performance for gastric cancer recurrence.
• Key risk factors identified include tumor invasion depth, lymph node metastasis, H. pylori infection, CEA, tumor size, and number.
• H. pylori infection was identified as the primary high-risk factor.

Conclusions:

• The XGBoost algorithm excels in predicting gastric cancer recurrence.
• Machine learning models can identify clinically significant prognostic factors for personalized monitoring and immunotherapy.