Clinical and translational medicine · 2026
Artificial intelligence and oral microbiome: Reshaping the diagnostic and therapeutic paradigm of OSCC.
СтатьяОбзорПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Shi Rui, Zhi Yuan, Gao Ling, Li Shao-Ming, Zhi Ke-Qian, Ren Wen-Hao
Аннотация
BACKGROUND: Oral squamous cell carcinoma (OSCC) remains a major clinical challenge, with delayed diagnosis, frequent resistance to therapy, and poor long-term survival.
METHODS: This review systematically evaluates the methodological framework for applying AI to oral microbiome data in OSCC. Emerging paradigms, including self-supervised learning for leveraging unlabelled data and explainable AI (XAI) techniques for model interpretability, are also discussed. Model evaluation relies on cross-validation, hyperparameter optimisation, and performance metrics such as AUC, accuracy, sensitivity, specificity, and F1-score.
RESULTS: Multiple studies demonstrate that AI-based classifiers, especially random forest models built on salivary or tissue-derived microbial features, achieve outstanding discrimination between OSCC patients and healthy controls in retrospective, single-centre cohorts, with reported AUC values exceeding 0.99 and accuracy >95%. However, these exceptional metrics should be interpreted with caution, as they are susceptible to cohort size, sampling site heterogeneity, batch effects, feature-selection bias, and the absence of independent external validation. Beyond binary diagnosis, AI has been successfully applied to predict lymph node metastasis, explore tumour metabolic reprogramming, and assess environmental interactions. Integrated multi-omics approaches further enhance the specificity and clinical relevance of microbial biomarkers.
CONCLUSIONS: The convergence of AI and oral microbiome analysis is reshaping the diagnostic and therapeutic landscape of OSCC, and explore microbiome-targeted combination therapies. Addressing these challenges will be pivotal to realising truly intelligent, personalised management and ultimately improving outcomes for OSCC patients.
METHODS: This review systematically evaluates the methodological framework for applying AI to oral microbiome data in OSCC. Emerging paradigms, including self-supervised learning for leveraging unlabelled data and explainable AI (XAI) techniques for model interpretability, are also discussed. Model evaluation relies on cross-validation, hyperparameter optimisation, and performance metrics such as AUC, accuracy, sensitivity, specificity, and F1-score.
RESULTS: Multiple studies demonstrate that AI-based classifiers, especially random forest models built on salivary or tissue-derived microbial features, achieve outstanding discrimination between OSCC patients and healthy controls in retrospective, single-centre cohorts, with reported AUC values exceeding 0.99 and accuracy >95%. However, these exceptional metrics should be interpreted with caution, as they are susceptible to cohort size, sampling site heterogeneity, batch effects, feature-selection bias, and the absence of independent external validation. Beyond binary diagnosis, AI has been successfully applied to predict lymph node metastasis, explore tumour metabolic reprogramming, and assess environmental interactions. Integrated multi-omics approaches further enhance the specificity and clinical relevance of microbial biomarkers.
CONCLUSIONS: The convergence of AI and oral microbiome analysis is reshaping the diagnostic and therapeutic landscape of OSCC, and explore microbiome-targeted combination therapies. Addressing these challenges will be pivotal to realising truly intelligent, personalised management and ultimately improving outcomes for OSCC patients.