Di Fede Olga, La Mantia Gaetano, Parola Marco, Maniscalco Laura, Matranga Domenica, Tozzo Pietro, Campisi Giuseppina, Cimino Mario G C A
Department of Precision Medicine in Medical, Surgical and Critical Care (Me.Pre.C.C.), University of Palermo, Palermo, Italy.
Unit of Oral Medicine and Dentistry for Fragile Patients, Department of Rehabilitation, Fragility, and Continuity of Care, University Hospital Palermo, Palermo, Italy.
Oral Dis. 2025 Apr;31(4):1054-1064. doi: 10.1111/odi.15188. Epub 2024 Nov 3.
Oral diseases, specifically malignant lesions, are serious global health concerns requiring early diagnosis for effective treatment. In recent years, deep learning (DL) has emerged as a powerful tool for the automated detection and classification of oral lesions. This research, by conducting a scoping review and meta-analysis, aims to provide an overview of the progress and achievements in the field of automated detection of oral lesions using DL.
A scoping review was conducted to identify relevant studies published in the last 5 years (2018-2023). A comprehensive search was conducted using several electronic databases, including PubMed, Web of Science, and Scopus. Two reviewers independently assessed the studies for eligibility and extracted data using a standardized form, and a meta-analysis was conducted to synthesize the findings.
Fourteen studies utilizing various DL algorithms were identified and included for the detection and classification of oral lesions from clinical images. Among these, three were included in the meta-analysis. The estimated pooled sensitivity and specificity were 0.86 (95% confidence interval [CI] = 0.80-0.91) and 0.67 (95% CI = 0.58-0.75), respectively.
The results of meta-analysis indicate that DL algorithms improve the diagnosis of oral lesions. Future research should develop validated algorithms for automated diagnosis.
Open Science Framework (https://osf.io/4n8sm).
口腔疾病，特别是恶性病变，是严重的全球健康问题，需要早期诊断以进行有效治疗。近年来，深度学习（DL）已成为用于口腔病变自动检测和分类的强大工具。本研究通过进行范围综述和荟萃分析，旨在概述使用DL进行口腔病变自动检测领域的进展和成果。
进行范围综述以识别过去5年（2018 - 2023年）发表的相关研究。使用包括PubMed、科学网和Scopus在内的多个电子数据库进行全面检索。两名评审员独立评估研究的 eligibility 并使用标准化表格提取数据，然后进行荟萃分析以综合研究结果。
确定了14项利用各种DL算法从临床图像中检测和分类口腔病变的研究并纳入其中。其中，3项纳入了荟萃分析。估计的合并灵敏度和特异度分别为0.86（95%置信区间[CI]=0.80 - 0.91）和0.67（95%CI = 0.58 - 0.75）。
荟萃分析结果表明DL算法改善了口腔病变的诊断。未来研究应开发经过验证的自动诊断算法。
开放科学框架（https://osf.io/4n8sm）。
