Li Ze, Zhao Kai, Liu Hongyu, Liu Jialin, Chen Xu, Hu Wentao, Wen Er, Zhang Kai, Chen Ling
Department of Neurosurgery, First Medical Center of the Chinese PLA General Hospital, Beijing, People's Republic of China.
China Medical University, Shenyang, People's Republic of China.
Brain Behav. 2025 Sep;15(9):e70843. doi: 10.1002/brb3.70843.
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.
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.
The AUC-ROC for the GBM model was 0.79, indicating a good level of discriminative ability.
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.
肠道微生物群在胶质瘤的发展中起着至关重要的作用。随着人工智能技术的发展，应用人工智能分析来自肠道微生物组的大量数据表明了人工智能和计算生物学在变革医学诊断和个性化医疗方面的潜力。
我们对42例术后诊断为胶质瘤的患者和30例非颅内肿瘤患者的粪便样本进行了宏基因组测序，并开发了一种梯度提升机（GBM）机器学习模型，以基于肠道微生物组数据预测胶质瘤患者。
GBM模型的AUC-ROC为0.79，表明具有良好的判别能力。
该方法在区分胶质瘤细胞与正常对照方面的有效性强调了机器学习模型利用大型数据集获取临床见解的潜力。
