Introduction: Polyetheretherketone (PEEK) is a polymer that is used in the construction of orthopaedic and dental implants. It is also used to construct removable and fixed dental prostheses due to its superior mechanical and esthetic properties compared to conventional materials. This systematic review aims to analyse and appraise the literature concerning PEEK dental prostheses critically. Methods: The following focused question was constructed ‘Are dental prostheses made of PEEK inferior to those made of other materials in terms of clinical- and patient-reported outcomes?’. The CONSORT (Consolidated Standards of Reporting Trials) tool was used for the quality assessment of the randomised clinical trials. The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) quality assessment tool was used to assess the quality of observational studies and the case reports were evaluated using the CARE (Case Report) guidelines. Results: A total of 12 studies were included in this review. Two case studies received an overall grade of medium and the overall quality of six studies was graded as ‘low’. All three observational studies and the only randomised controlled trial received scores of ‘medium’. Conclusion: PEEK-based dental prostheses may provide a viable and more esthetic alternative to conventional prosthodontic appliances. However, within the limitations of this study is the evidence to ascertain the long-term viability of PEEK-based dental prostheses. Future studies should focus on conducting large-scale, multicenter trials to compare the survival rate of PEEK prostheses to that of conventionally available prosthodontic appliances.

Automated prediction of dental conditions in Orthopantomogram (OPG) panoramic radiographs faces significant challenges due to class imbalance, rare pathologies, and complex anatomical structures. This study proposes OralHybridNet, a novel hybrid deep learning framework integrating hierarchical convolutional neural networks, such as CustomDentalNet integrate dual-attention mechanisms and OralNetXPlus. A multinational dataset comprising 2047 clinician-annotated panoramic radiographs spanning 7 diagnostic labels was used. An adaptive augmentation protocol combining Elastic Transformations and gamma correction mitigated class imbalance. A Hybrid Feature Selection (HFS) algorithm condensed 1208-dimensional embeddings into a discriminative 300-feature subset. Evaluated against ResNet50 baselines, OralHybridNet achieved 96.0% accuracy, 97.6% precision, and 0.993 AUC-ROC. The KNN Fine classifier on fused features yielded the highest performance, with real-time capability (9 ms inference time) using a neural network classifier. The proposed framework demonstrates a promising proof-of-concept framework for automated multi-label dental restoration classification.

Aim: This randomized controlled study aimed to evaluate clinical outcomes of bar versus resilient stud (locator) anchors used for immediately loaded 2 implants assisting overdentures in patients with history of periodontitis Materials and methods: Twelve completely edentulous patients with a history of chronic periodontitis were randomly classified into 2 groups; Group 1 (locator group); included six participants who received resilient stud (Locator) attachments for overdentures. Group 2 (bar group); included six participants who received bar attachments for overdentures. All patients received 2 implants in the cuspid regions of the mandible using computer guided flapless surgical approach and the implants were immediately loaded with overdentures. Clinical outcomes (implant survival, plaque scores, gingival scores, probing depth, stability of implants, width of keratinized mucosa, and crestal bone loss) were measured at overdenture insertion, 6 and 12 months later. Results: Plaque and gingival indices, pocket depth, and crestal bone loss increased significantly with time in both groups. Implant stability significantly decreased in locator group and significantly increased with bar group with advance of time. Keratinized mucosal width significantly diminished with time for both attachments. Bar group recorded significant higher plaque and gingival indices, and stability of implants than locator group. On the other hand, Locator group recorded recoded significant higher marginal bone loss than bar group. Conclusion: Within the limits of this study, Bar attachments are recommended to assist overdentures to immediate loaded implants in subjects with history of periodontitis. However, it is accompanied by increased plaque and gingival scores than Locator attachment.

Artificial intelligence (AI) has emerged as a promising tool in oral and maxillofacial surgery (OMFS), particularly in enhancing diagnostic accuracy, surgical planning, and risk prediction. This narrative review critically evaluates current AI applications across key clinical domains, including lesion detection and classification in diagnostic imaging, prediction of surgical difficulty and inferior alveolar nerve injury in impacted third molar surgery, cone-beam computed tomography–based bone assessment and implant planning, automated landmark detection and surgical simulation in orthognathic surgery, classification of temporomandibular joint disorders, and prognostic modeling in oral oncology. Despite encouraging results, most studies are based on retrospective, single-center datasets with limited external or prospective validation. Methodological heterogeneity, inconsistent outcome definitions, and insufficient integration into clinical workflows remain major barriers to translation. While AI demonstrates significant potential as a decision-support tool in OMFS, current evidence does not yet support widespread clinical implementation. Future research should prioritize robust validation strategies, clinically meaningful endpoints, and seamless integration into surgical practice.