Non-Cochrane · DTA · PM42333867
Artificial Intelligence-Driven MRI for Cervical Nodal Metastasis Detection in Oral Squamous Cell Carcinoma: A Hierarchical Meta-Analysis of Diagnostic Accuracy.
da Silva-Filho José Evando, Ribeiro Renata Roque, de Aguiar André Wescley Oliveira, et al. · Head & neck · 2026 · PubMed · doi:10.1002/hed.70367
Findings
18 findings
| Population | ||||
|---|---|---|---|---|
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Leave-one-out sensitivity analysis: Troise et al. (2025) omitted | Specificity 0.79 GRADE not reported 6 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis | Sensitivity 0.72 [0.62, 0.80] Low Downgraded: risk of bias, indirectness 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis | Specificity 0.79 [0.73, 0.83] Low Downgraded: risk of bias, indirectness 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Sensitivity analysis: abstract-reported values substituted for Troise et al. (2025) | Specificity 0.79 GRADE not reported 7 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Area under the HSROC curve Preoperative staging of cervical lymph node metastasis | AUC (HSROC) 0.82 GRADE not reported 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Diagnostic odds ratio Preoperative staging of cervical lymph node metastasis | DOR 9.42 [6.65, 13.54] GRADE not reported 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Sensitivity analysis: abstract-reported values substituted for Troise et al. (2025) | Sensitivity 0.71 GRADE not reported 7 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Negative likelihood ratio Preoperative staging of cervical lymph node metastasis | LR- 0.36 [0.28, 0.47] GRADE not reported 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Leave-one-out sensitivity analysis: Troise et al. (2025) omitted | Sensitivity 0.725 GRADE not reported 6 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Positive likelihood ratio Preoperative staging of cervical lymph node metastasis | LR+ 3.23 [2.67, 4.18] GRADE not reported 7 studies · 548 participants |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based MRI algorithms, patient-level analyses | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Patient-level studies (unit of analysis) | Sensitivity 0.706 GRADE not reported 5 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | AI-based MRI algorithms, patient-level analyses | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Patient-level studies (unit of analysis) | Specificity 0.776 GRADE not reported 5 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI (mixed-stage cohorts) | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Mixed-stage cohorts (clinical stage) | Sensitivity 0.741 GRADE not reported 5 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI (mixed-stage cohorts) | AI-based algorithms (radiomics machine learning or deep learning) applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Mixed-stage cohorts (clinical stage) | Specificity 0.777 GRADE not reported 5 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | Radiomics-based machine learning models applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Radiomics-based machine learning models | Sensitivity 0.667 GRADE not reported 4 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | Radiomics-based machine learning models applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Radiomics-based machine learning models | Specificity 0.801 GRADE not reported 4 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | Deep learning models applied to MRI | Histopathology reference standard (cervical lymph node status) | Specificity Preoperative staging of cervical lymph node metastasis · Deep learning models | Specificity 0.783 GRADE not reported 3 studies |
| Adults with histopathologically confirmed oral squamous cell carcinoma undergoing MRI | Deep learning models applied to MRI | Histopathology reference standard (cervical lymph node status) | Sensitivity Preoperative staging of cervical lymph node metastasis · Deep learning models | Sensitivity 0.738 GRADE not reported 3 studies |
Included primary studies 12
Open in Studies| Study | Year | Also included by | Outcomes it informs here | Trial data |
|---|---|---|---|---|
| Wang 2022 Magnetic Resonance Imaging-Based Radiomics Features Associated with Depth of Invasion Predicted Lymph Node Metastasis and Prognosis in… PMID:34888985 | 2022 | 2 other reviews | not stated | not extracted yet |
| Yuan 2021 Machine learning-based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma. PMID:33569617 | 2021 | 2 other reviews | Sensitivity | not extracted yet |
| Lan 2024 MRI-based deep learning and radiomics for prediction of occult cervical lymph node metastasis and prognosis in early-stage oral and… PMID:38729119 | 2024 | 1 other review | not stated | not extracted yet |
| Liu 2024 The application of radiomics machine learning models based on multimodal MRI with different sequence combinations in predicting cervical… PMID:38108536 | 2024 | 1 other review | Sensitivity | not extracted yet |
| Ren 2022 Histogram analysis of diffusion-weighted imaging and dynamic contrast-enhanced MRI for predicting occult lymph node metastasis in… PMID:34642806 | 2022 | 1 other review | not stated | not extracted yet |
| Traverso 2020 matched by author and year (probable) | NR | 1 other review | not stated | not extracted yet |
| Wang 2024 Magnetic resonance imaging-based radiomics and deep learning models for predicting lymph node metastasis of squamous cell carcinoma of the… PMID:38378316 | 2024 | 1 other review | Sensitivity | not extracted yet |
| Gang 2024 Predicting nodal metastasis progression of oral tongue cancer using a hidden Markov model in MRI. PMID:38912064 | 2024 | this review only | not stated | not extracted yet |
| Ho 2020 Classifying Neck Lymph Nodes of Head and Neck Squamous Cell Carcinoma in MRI Images with Radiomic Features. PMID:31950301 | 2020 | this review only | Sensitivity | not extracted yet |
| Troise 2025 The Role of Machine Learning to Detect Occult Neck Lymph Node Metastases in Early-Stage (T1-T2/N0) Oral Cavity Carcinomas. PMID:40390252 | 2025 | this review only | Sensitivity | not extracted yet |
| Yang 2025 Diagnosis of lymph node metastasis in oral squamous cell carcinoma by an MRI-based deep learning model. PMID:39752793 | 2025 | this review only | Sensitivity | not extracted yet |
| Zheng 2025 Application value of dual-sequence MRI based nomogram of radiomics and morphologic features in predicting tumor differentiation degree and… PMID:40735034 | 2025 | this review only | Sensitivity | not extracted yet |
Overlap with other reviews
Reviews that include the same primary studies are not independent evidence. Shared counts match studies by PMID, DOI or NCT, and otherwise by author and year (a probable match). Only the non-Cochrane dental reviews have study lists, so overlap with Cochrane reviews is not counted yet.
Review question
| Element | Planned (methods) | Found (results) |
|---|---|---|
| Population | Adults (18 years or older) with histopathologically confirmed OSCC who underwent head and neck MRI | OSCC patients, mainly oral cavity and tongue (12 studies; 548 participants in 7 meta-analyzed datasets) |
| Intervention | AI-based algorithms (fully automated or radiologist-assisted) applied to MRI | AI-based MRI: radiomics with machine learning, deep learning, hidden Markov model (n=12) |
| Comparison | Histopathological reference standard | Histopathology (n=12) |
| Outcomes | Sensitivity, specificity, AUC, accuracy, Dice similarity coefficient, IoU, LR+, LR-, DOR for cervical lymph node detection, segmentation and metastasis identification | Pooled sensitivity 0.72, specificity 0.79, AUC 0.82, LR+ 3.23, LR- 0.36, DOR 9.42 (n=7) |
| Study designs | Prospective or retrospective observational studies from 2000 | Retrospective observational studies (n=12) |
| Timing | Not specified | Not specified |