Multi-strategy feature selection and multi-model machine learning for prognostic prediction in primary gastric diffuse large B-cell lymphoma

1

Primary gastric diffuse large B-cell lymphoma (PG-DLBCL) is the most common subtype of primary gastric lymphoma, with heterogeneous prognosis despite improved treatments.
2

Conventional prognostic systems like the International Prognostic Index may not accurately assess PG-DLBCL due to its unique clinicopathological features.
3

The study aims to create a robust prognostic model for PG-DLBCL using SEER data, integrating multi-strategy variable selection with machine learning algorithms.
4

Four complementary feature selection strategies were employed to identify key predictors, enhancing model interpretability and minimizing overfitting risk.
5

The research utilizes multiple machine learning frameworks to develop a prognostic tool, aiming to improve risk stratification and individualized patient management.

Digital Health

1