The diagnostic value of artificial intelligence in oral squamous cell carcinoma: A systematic review and meta-analysis

Cong Ren ¹, Chengfeng Wang ², Ren Lin ²

⁰Lishui University, Lishui City 323000 Zhejiang Province, China; Lishui Stomatological Hospital, Lishui City 323000 Zhejiang Province, China.

Journal of Stomatology, Oral and Maxillofacial Surgery +

|

June 15, 2025

View abstract on PubMed

Summary

Artificial intelligence (AI) shows high accuracy in detecting oral squamous cell carcinoma (OSCC). This systematic review suggests AI

Area Of Science

Medical Informatics
Oncology
Diagnostic Imaging

Background

Oral squamous cell carcinoma (OSCC) is a significant global health concern.
Early and accurate diagnosis is crucial for improving patient outcomes.
Current diagnostic methods face challenges in sensitivity and specificity.

Purpose Of The Study

To systematically review and meta-analyze the diagnostic performance of artificial intelligence (AI) for OSCC detection.
To quantify the accuracy metrics of AI algorithms in identifying OSCC.
To compare the performance of different AI approaches in OSCC diagnosis.

Main Methods

A comprehensive literature search was performed across major databases (PubMed, Scopus, Web of Science) from 2000 to 2023.
Studies evaluating AI for OSCC diagnosis with sufficient data for accuracy calculation were included.
A bivariate random-effects model was employed to pool diagnostic performance metrics, including sensitivity, specificity, PLR, NLR, and DOR. Methodological quality was assessed using QUADAS-2.

Main Results

Twenty-four studies involving 18,574 specimens were analyzed.
Pooled sensitivity was 0.95 (95% CI: 0.90-0.98) and pooled specificity was 0.95 (95% CI: 0.91-0.98).
Deep learning algorithms outperformed conventional machine learning methods, despite significant heterogeneity observed across studies (I² > 97%).

Conclusions

AI demonstrates high diagnostic accuracy for OSCC detection, positioning it as a potential adjunctive tool in clinical settings.
The significant heterogeneity highlights the need for standardized methodologies and external validation prior to widespread clinical adoption.
Further research focusing on standardized protocols and validation is recommended.

Keywords:

Artificial intelligence Deep learning Diagnostic accuracy Machine learning Meta-analysis Oral squamous cell carcinoma