Construction of Predictive Machine Learning Model of Glioma-Associated Gut Microbiota.

BACKGROUND

The gut microbiota plays a crucial role in the development of glioma. With the evolution of artificial intelligence technology, applying AI to analyze the vast amount of data from the gut microbiome indicates the potential that artificial intelligence and computational biology hold in transforming medical diagnostics and personalized medicine.

METHODS

We conducted metagenomic sequencing on stool samples from 42 patients diagnosed with glioma after operation and 30 non-intracranial tumor patients and developed a Gradient Boosting Machine (GBM) machine learning model to predict the glioma patients based on the gut microbiome data.

RESULTS

The AUC-ROC for the GBM model was 0.79, indicating a good level of discriminative ability.

CONCLUSIONS

This method's efficacy in discriminating between glioma cells and normal controls underscores the potential of machine learning models in leveraging large datasets for clinical insights.