BMC oral health · 2026
Application of artificial intelligence and radiomics in the prediction of lymph node metastasis and tumour grading of oral cancer - a systematic review and meta analysis.
СтатьяСистематический обзорМетаанализНазвание 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.
Mohideen Khadijah, Ghosh Snehashish, Krithika Chandrasekaran, Mulk Bhavana Sujana, Chole Revant, Chatterjee Juhi, Dhungel Safal
Аннотация
BACKGROUND: Radiomics investigation strategies can be applied to head and neck tumours, including lesion segmentation, tumour grading and staging prediction. Texture features from PET/CT radiomics, particularly those reflecting metabolic heterogeneity within the primary tumour, have shown substantial predictive value for lymph node metastasis in oral cancer. Accurate prediction of cervical lymph node metastasis in oral cancer is crucial, as it is the most significant prognostic factor influencing treatment planning and patient survival.
METHOD: An extensive search across PubMed, Scopus, and Wiley Online Library, adhering to PRISMA guidelines, was carried out. The present review included 40 studies, of which 33 were included in the meta-analysis of the prediction of lymph node metastasis and tumour grading.
RESULTS: The pooled sensitivity, specificity and Diagnostic Odds Ratio (DOR) of the AI models for the prediction of LN metastases were 0.86 (95% CI 0.80-0.90), 0.91 (95% CI 0.87-0.93), and 56.58 (95% CI 21.68-91.48), respectively. The pooled sensitivity, specificity and DOR of the AI models for the grading of OSCC were 0.88 (95% CI 0.54-0.98), 0.82 (95% CI 0.76-0.87), and 34.38 (95% CI 24.24-103), respectively.
CONCLUSION: To mitigate the elevated misinterpretation rate of lymph node metastasis (LNMs), it is prudent to incorporate ML/DL into the imaging identification of LNMs in oral cancer. Radiomic CT characteristics of oral cancer indicate tumour heterogeneity and can forecast histopathologic attributes. These exploratory investigations suggest that the AI and radiomics prediction framework may function as an additional non-invasive diagnostic tool for oral cancer, enhancing the objectivity and accuracy of tumour staging and grading and providing guidance for future therapies.
METHOD: An extensive search across PubMed, Scopus, and Wiley Online Library, adhering to PRISMA guidelines, was carried out. The present review included 40 studies, of which 33 were included in the meta-analysis of the prediction of lymph node metastasis and tumour grading.
RESULTS: The pooled sensitivity, specificity and Diagnostic Odds Ratio (DOR) of the AI models for the prediction of LN metastases were 0.86 (95% CI 0.80-0.90), 0.91 (95% CI 0.87-0.93), and 56.58 (95% CI 21.68-91.48), respectively. The pooled sensitivity, specificity and DOR of the AI models for the grading of OSCC were 0.88 (95% CI 0.54-0.98), 0.82 (95% CI 0.76-0.87), and 34.38 (95% CI 24.24-103), respectively.
CONCLUSION: To mitigate the elevated misinterpretation rate of lymph node metastasis (LNMs), it is prudent to incorporate ML/DL into the imaging identification of LNMs in oral cancer. Radiomic CT characteristics of oral cancer indicate tumour heterogeneity and can forecast histopathologic attributes. These exploratory investigations suggest that the AI and radiomics prediction framework may function as an additional non-invasive diagnostic tool for oral cancer, enhancing the objectivity and accuracy of tumour staging and grading and providing guidance for future therapies.