Non-Cochrane · DTA;prognostic · PM42185818
Artificial intelligence for oral cancer diagnosis: a systematic review and meta-analysis of image-based and non-imaging models.
Soleimani Sardou Shima, Rezvaninejad Raziyehsadat, Rezvaninejad Fatemeh Sadat, et al. · BMC cancer · 2026 · PubMed · doi:10.1186/s12885-026-16154-4
Findings
Population Patients or samples with oral lesions, oral potentially malignant disorders or oral cancer
9 findings
| AI models for oral cancer diagnosis or risk assessment (image-based and text-based) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic accuracy Diagnostic and risk assessment models for oral cancer · Overall (all data types) | Pooled accuracy (proportion) 0.89 [0.83, 0.95] GRADE not reported 11 studies |
| AI models for oral cancer diagnosis or risk assessment (image-based and text-based) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic sensitivity Diagnostic and risk assessment models for oral cancer · Overall (all data types) | Pooled sensitivity (proportion) 0.90 [0.81, 0.98] GRADE not reported 13 studies |
| AI models for oral cancer diagnosis or risk assessment (image-based and text-based) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic specificity Diagnostic and risk assessment models for oral cancer · Overall (all data types) | Pooled specificity (proportion) 0.89 [0.84, 0.95] GRADE not reported 14 studies |
| Text-based (non-imaging) AI models using clinical, questionnaire or record variables | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic specificity Diagnostic and risk assessment models for oral cancer · Text-based (non-imaging) models | Pooled specificity (proportion) 0.79 [0.67, 0.91] GRADE not reported 4 studies |
| Text-based (non-imaging) AI models using clinical, questionnaire or record variables | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic sensitivity Diagnostic and risk assessment models for oral cancer · Text-based (non-imaging) models | Pooled sensitivity (proportion) 0.76 [0.42, 1.10] GRADE not reported 3 studies |
| Text-based (non-imaging) AI models using clinical, questionnaire or record variables | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic accuracy Diagnostic and risk assessment models for oral cancer · Text-based (non-imaging) models | Pooled accuracy (proportion) 0.81 [0.70, 0.92] GRADE not reported 4 studies |
| Image-based AI models (clinical photographs, histopathology, hyperspectral, OCT, other biomedical images; one saliva-based study grouped here) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic specificity Diagnostic and risk assessment models for oral cancer · Image-based models | Pooled specificity (proportion) 0.93 [0.90, 0.97] GRADE not reported 10 studies |
| Image-based AI models (clinical photographs, histopathology, hyperspectral, OCT, other biomedical images; one saliva-based study grouped here) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic accuracy Diagnostic and risk assessment models for oral cancer · Image-based models | Pooled accuracy (proportion) 0.93 [0.89, 0.97] GRADE not reported 7 studies |
| Image-based AI models (clinical photographs, histopathology, hyperspectral, OCT, other biomedical images; one saliva-based study grouped here) | Reference standard (histopathology where available, otherwise expert clinical diagnosis) | Diagnostic sensitivity Diagnostic and risk assessment models for oral cancer · Image-based models | Pooled sensitivity (proportion) 0.94 [0.90, 0.97] GRADE not reported 10 studies |
Included primary studies 28
Open in Studies| Study | Year | Also included by | Outcomes it informs here | Trial data |
|---|---|---|---|---|
| Adeoye 2023 Drug use and HIV-1 infection: report from the Second Italian Multicenter Study. PMID:2023101 | 1991 | 1 other review | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Ahmad 2023 Multi-Method Analysis of Histopathological Image for Early Diagnosis of Oral Squamous Cell Carcinoma Using Deep Learning and Hybrid… PMID:37958422 | 2023 | 1 other review | not stated | not extracted yet |
| Fu 2020 A deep learning algorithm for detection of oral cavity squamous cell carcinoma from photographic images: A retrospective study DOI:10.1016/j.eclinm.2020.100558 | 2020 | 1 other review | Diagnostic sensitivity; Diagnostic specificity | not extracted yet |
| Fu 2020 A deep learning algorithm for detection of oral cavity squamous cell carcinoma from photographic images: A retrospective study. PMID:33150326 | 2020 | 1 other review | not stated | not extracted yet |
| Jeyaraj 2019 Computer-assisted medical image classification for early diagnosis of oral cancer employing deep learning algorithm. PMID:30603908 | 2019 | 1 other review | not stated | not extracted yet |
| Jubair 2021 A novel lightweight deep convolutional neural network for early detection of oral cancer DOI:10.1111/odi.13825 | 2021 | 1 other review | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Jubair 2022 A novel lightweight deep convolutional neural network for early detection of oral cancer. PMID:33636041 | 2022 | 1 other review | not stated | not extracted yet |
| Warin 2021 Automatic classification and detection of oral cancer in photographic images using deep learning algorithms. PMID:34358372 | 2021 | 1 other review | not stated | not extracted yet |
| Warin 2022 Performance of deep convolutional neural network for classification and detection of oral potentially malignant disorders in photographic… DOI:10.1016/j.ijom.2021.09.001 | 2022 | 1 other review | Diagnostic sensitivity; Diagnostic specificity | not extracted yet |
| Ahmad 2023 Multi-Method Analysis of Histopathological Image for Early Diagnosis of Oral Squamous Cell Carcinoma Using Deep Learning and Hybrid… DOI:10.3390/cancers15215247 | 2023 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Alhazmi 2021 Application of artificial intelligence and machine learning for prediction of oral cancer risk DOI:10.1111/jop.13157 | 2021 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Alhazmi 2021 Application of artificial intelligence and machine learning for prediction of oral cancer risk. PMID:33394536 | 2021 | this review only | not stated | not extracted yet |
| Banavar 2021 The salivary metatranscriptome as an accurate diagnostic indicator of oral cancer DOI:10.1038/s41525-021-00257-x | 2021 | this review only | Diagnostic sensitivity; Diagnostic specificity | not extracted yet |
| Banavar 2021 The salivary metatranscriptome as an accurate diagnostic indicator of oral cancer. PMID:34880265 | 2021 | this review only | not stated | not extracted yet |
| Chen 2024 Intelligent deep learning supports biomedical image detection and classification of oral cancer DOI:10.3233/thc-248041 | 2024 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Chen 2024 Intelligent deep learning supports biomedical image detection and classification of oral cancer. PMID:38759069 | 2024 | this review only | not stated | not extracted yet |
| Chu 2020 Machine learning and treatment outcome prediction for oral cancer DOI:10.1111/jop.13089 | 2020 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Chu 2020 Machine learning and treatment outcome prediction for oral cancer. PMID:32740951 | 2020 | this review only | not stated | not extracted yet |
| Jeyaraj 2019 Computer-assisted medical image classification for early diagnosis of oral cancer employing deep learning algorithm DOI:10.1007/s00432-018-02834-7 | 2019 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Panigrahi 2023 Classifying histopathological images of oral squamous cell carcinoma using deep transfer learning DOI:10.1016/j.heliyon.2023.e13444 | 2023 | this review only | Diagnostic accuracy | not extracted yet |
| Panigrahi 2023 Classifying histopathological images of oral squamous cell carcinoma using deep transfer learning. PMID:37101475 | 2023 | this review only | not stated | not extracted yet |
| Shimpi 2022 Development and Validation of a Non-Invasive, Chairside Oral Cavity Cancer Risk Assessment Prototype Using Machine Learning Approach. PMID:35455730 | 2022 | this review only | Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Shree 2024 Saliva Based Diagnostic Prediction of Oral Squamous Cell Carcinoma using FTIR Spectroscopy DOI:10.1007/s12070-023-04294-z | 2024 | this review only | not stated | not extracted yet |
| Shree 2024 Saliva Based Diagnostic Prediction of Oral Squamous Cell Carcinoma using FTIR Spectroscopy. PMID:38883442 | 2024 | this review only | not stated | not extracted yet |
| Warin 2021 Automatic classification and detection of oral cancer in photographic images using deep learning algorithms DOI:10.1111/jop.13227 | 2021 | this review only | Diagnostic sensitivity; Diagnostic specificity | not extracted yet |
| Warin 2022 Performance of deep convolutional neural network for classification and detection of oral potentially malignant disorders in photographic… PMID:34548194 | 2022 | this review only | not stated | not extracted yet |
| Yuan 2023 Noninvasive diagnosis of oral squamous cell carcinoma by multi‐level deep residual learning on optical coherence tomography images DOI:10.1111/odi.14318 | 2022 | this review only | Diagnostic sensitivity; Diagnostic specificity; Diagnostic accuracy | not extracted yet |
| Yuan 2023 Noninvasive diagnosis of oral squamous cell carcinoma by multi-level deep residual learning on optical coherence tomography images. PMID:35842738 | 2023 | this review only | not stated | 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 | Studies of AI-assisted screening or diagnosis of oral mucosal lesions and oral cancer | Oral cancer and oral lesion datasets (sample sizes 60 to 44,409) |
| Intervention | Artificial neural networks, machine learning, deep learning, expert systems, clinical decision support systems | AI models: text-based (n=4 studies), image-based (n=12 studies) |
| Comparison | Accepted reference standard (histopathology where available, or expert clinical diagnosis) | Histology, expert clinical diagnosis, or not clearly stated |
| Outcomes | Sensitivity, specificity, accuracy, AUC; prognostic outcomes extracted when reported | Pooled sensitivity (n=13), specificity (n=14), accuracy (n=11); AUC and prognostic outcomes narrative only |
| Study designs | Original prospective or retrospective studies, English, full text, January 2010 to September 2024 | Mostly retrospective diagnostic accuracy studies (n=16) |
| Timing | Not specified | Not specified |