Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 5 appraisals
Dento maxillo facial radiology
JADE: jawbone lesion diagnosis and decision supporting system
OBJECTIVES: To develop and evaluate JADE, a proof-of-concept retrieval-augmented generation (RAG) diagnostic assistive system, designed to enhance large language model (LLM) reasoning for jawbone lesion assessment. This study examined whether RAG improves diagnostic accuracy and stability compared with standalone LLMs and ORAD, a supervised learning-based system. METHODS: JADE was developed as a cloud-based RAG system integrating an expert-curated oral radiology database embedded using text-embedding-3-large and indexed in Qdrant Cloud. Structured clinical inputs were encoded as prioritized vector queries. Hybrid semantic and keyword-based retrieval appended relevant evidence to prompts for differential diagnosis generation across LLM backbones. Performance was evaluated in 25 validation cases and compared with GPT-5, Claude Sonnet 4.5, DeepSeek-R1, Gemini 2.5 Flash, their RAG configurations, and ORAD. Diagnostic accuracy was analyzed using Cochran's Q test, with post-hoc McNemar's tests and Bonferroni correction. Intra-model stability and response time were assessed. RESULTS: RAG-GPT-5 achieved the highest diagnostic accuracy (20/25), followed by RAG-Claude Sonnet 4.5 (18/25), RAG-DeepSeek R1 (17/25), and RAG-Gemini 2.5 Flash (15/25). Standalone models achieved 9-13/25 correct diagnoses, ORAD achieved 17/25. No significant differences were observed among standalone models or RAG-based models and ORAD. A significant improvement was observed for GPT-5 when integrated with RAG (P = .002). RAG configurations showed higher intra-model stability, with RAG-GPT-5 achieving mean stability of 0.90 ± 0.11. Mean response times ranged from 3 to 10 s. CONCLUSIONS: JADE improved diagnostic accuracy and stability compared with standalone LLMs, underscoring value of RAG reasoning in jawbone lesion assessment and marking the first RAG application in dentomaxillofacial radiology.
3 July 2026
Read appraisal →Molecular biology reports
Oral mucosal microbiome alterations in recurrent aphthous stomatitis: a systematic review of 16 S rRNA gene sequencing studies
Recurrent aphthous stomatitis (RAS) is a prevalent inflammatory disorder of the oral mucosa characterized by recurrent painful ulcerations in otherwise healthy individuals. This systematic review aimed to evaluate alterations in the oral mucosal microbiome of patients with RAS based on studies using 16 S rRNA sequencing. A systematic search of PubMed, Scopus, and Web of Science was conducted on April 14, 2026. Eligible studies included human case-control investigations evaluating oral mucosal swab samples from patients with clinically diagnosed RAS and healthy controls using 16 S rRNA sequencing. Studies based solely on saliva, culture methods, PCR-only analyses, or lacking controls were excluded. Joanna Briggs Institute Critical Appraisal Checklist for Case-Control Studies was used for the evaluation of selected articles. Six studies met the inclusion criteria. Considerable heterogeneity was observed in alpha and beta diversity outcomes. Most studies reported reduced microbial richness in RAS lesions, whereas others found increased or unchanged diversity. Ulcerated sites frequently demonstrated reduced abundance of health-associated taxa such as Streptococcus and Firmicutes, with increased levels of Proteobacteria and inflammation-associated genera including Neisseria, Haemophilus, Prevotella, and Fusobacterium. Microbial alterations were most pronounced at active ulcer sites, while non-ulcerated or healed mucosa more closely resembled healthy controls. Current evidence suggests that RAS is associated with localized, site-specific microbial dysbiosis rather than generalized oral microbiome disruption. However, methodological heterogeneity and small sample sizes limit definitive conclusions. Future standardized longitudinal studies integrating functional metagenomics are warranted to clarify the role of the microbiome in RAS pathogenesis.
16 May 2026
Read appraisal →Calcified tissue international
Diagnostic Indices for Epidemiological Assessment of Molar Incisor Hypomineralization: A Systematic Review
Molar incisor hypomineralization (MIH) is a qualitative developmental defect of enamel and one of the most prevalent oral health conditions in childhood. Over the last two decades, increasing attention has been given to its prevalence and etiology; however, marked heterogeneity in diagnostic criteria and indices has limited the comparability of epidemiological findings. This systematic review aimed to identify and describe the diagnostic indices used for the epidemiological assessment of MIH in the general pediatric population. A systematic literature search was conducted in PubMed, Scopus, and Google Scholar to identify epidemiological studies published between January 2001 and December 2024 reporting MIH prevalence in children. Study selection, data extraction, and methodological quality assessment were performed independently by two reviewers in accordance with PRISMA guidelines. The review protocol was registered in PROSPERO (CRD42022345224). A total of 198 articles were included, corresponding to 201 independent epidemiological studies, as three articles reported results from two distinct study populations. The European Academy of Paediatric Dentistry (EAPD) criteria were used in 171 studies (75.7%), representing the most widely adopted diagnostic index. Other indices, including the modified Developmental Defects of Enamel index (mDDE), Mathu-Muju and Wright, Wetzel and Reckel, MIH Severity Scoring System (MIH-SSS), the MIH Treatment Need Index (MIH TNI) criteria, and FDI criteria, were used infrequently. Studies using EAPD criteria accounted for the largest cumulative sample size and showed a descriptively aggregated MIH prevalence of 12.9%, whereas prevalence estimates varied across alternative diagnostic indices. In conclusion, the lack of uniform diagnostic standards and nomenclature hampers global surveillance and limits the comparability of prevalence data. The adoption of MIH-specific diagnostic criteria, supported by adequate examiner calibration and transparent reporting, is essential to improve the quality, consistency, and interpretability of future epidemiological research.
11 May 2026
Read appraisal →The International journal of oral & maxillofacial implants
Histologic and Immunohistochemical Analysis of Peri-implant Tissue Regeneration at the Interface of Prosthetic Abutments Treated with Anodizing: A Prospective, Randomized Controlled Clinical Trial on the Early Postoperative Period
PURPOSE: To assess the effects of anodized prosthetic abutments versus machined abutments on peri-implant connective tissue. MATERIALS AND METHODS: Following dental implant placement and osseointegration, custom healing abutments with machined surfaces (G1) and anodized surface treatment (G2) were attached to 28 dental implants. After a soft tissue healing period of approximately 30 days (± 7 days), 5-mm circular biopsy samples were taken from the peri-implant tissue and the area adjacent to the prosthetic abutment. Histologic analysis was conducted on the stained specimens using hematoxylin-eosin (H&E), focusing on parameters such as vascular proliferation as well as the presence of polymorphonuclear and mononuclear cells, collagen fibrils, and reepithelialization. A grading scale from 0 to 3 evaluated the extent of the inflammatory response. Masson's trichrome staining was used to analyze collagen fiber presence and organization, while immunohistochemical staining assessed the expression of beta-catenin (β-catenin) and Cluster of Differentiation 34 (CD34). The prosthetic abutments were evaluated through scanning electron microscopy and 3D roughness analysis. RESULTS: Histologic findings revealed that peri-implant tissue adjacent to anodized surfaces exhibited epithelial stratification and a parakeratin layer in the lamina propria. Both groups displayed a typical mononuclear lymphocytic inflammatory infiltrate, but the intensity was significantly higher adjacent to the machined abutment (P < .05). Quantification of blood vessels indicated increased immunopositivity for CD34 in G2 compared to G1 (P < .05). Immunolabeling of β-catenin was detected in the epithelial tissue, showing weak intensity in G1 and strong intensity in G2. The average surface roughness measurements were determined to be 0.097 ± 0.006 μm for G1 and 0.182 ± 0.008 μm for G2. CONCLUSIONS: Anodized surfaces are a promising treatment option and exhibit beneficial effects on peri-implant tissue healing. They play an important role in preserving the epithelial mucosal integrity and functionality near the prosthetic abutment, thereby potentially minimizing the risks of mucositis, peri-implantitis, and implant failure.
5 Apr 2026
Read appraisal →Journal of dentistry
A deep learning method for diagnosis of oral potentially malignant disorders
OBJECTIVES: This study aimed to develop and validate a two-stage deep learning method for diagnosing oral potentially malignant disorders (OPMDs). We also compared its diagnostic performance with that of clinicians at different levels of seniority and evaluated its utility as a clinical decision support tool. METHODS: A two-stage deep learning method was developed. CLA OPMD-OOML was designed to differentiate OPMDs from other oral mucosal lesions (OOML), while CLA OPMDs was designed to classify specific OPMD subtypes. The method was trained on an internal dataset (ZJUSS, n = 3305), and its generalizability was evaluated on two external multicenter datasets (WCHS and CS-SJTU, n = 1756). In a blinded, two-step comparative study, nine clinicians (junior, intermediate, and senior groups) performed diagnoses independently and then with AI assistance. RESULTS: The method significantly outperformed all clinician groups, achieving higher F1 scores in both CLA OPMD-OOML (89.9 % vs. 78.4-82.6 % for clinicians) and CLA OPMDs (92.2 % vs. 77.7-80.0 % for clinicians). It showed robust generalizability on the two external datasets, with F1 scores of 87.3 % and 86.9 % for CLA OPMD-OOML and 75.9-83.0 % for CLA OPMDs. With AI assistance, the diagnostic precision of junior and intermediate clinicians increased by 10.8 % and 5.9 %, respectively, raising the junior group's performance to the level of senior clinicians. CONCLUSIONS: The developed two-stage deep learning method demonstrated diagnostic performance comparable to or exceeding that of experienced clinicians in classifying OPMDs from clinical images. It functions as a powerful assistive tool that substantially enhances the diagnostic capabilities of junior and intermediate clinicians. CLINICAL SIGNIFICANCE: This AI method has the potential to serve as a reliable tool for large-scale early screening of OPMDs, particularly in regions with limited access to specialist dental care. It can also serve as a valuable training and decision support system for junior clinicians, helping to standardize diagnostic accuracy and improve patient outcomes through timely and accurate detection. A supplemental appendix to this article is available online.
4 Apr 2026
Read appraisal →