21 July 2022 in Journal of Dentistry
19 June 2024 in European Journal of Orthodontics
This systematic review and meta-analysis aimed to investigate the accuracy and efficiency of artificial intelligence (AI)-driven automated landmark detection for cephalometric analysis on two-dimensional (2D) lateral cephalograms and three-dimensional (3D) cone-beam computed tomographic (CBCT) images. An electronic search was conducted in the following databases: PubMed, Web of Science, Embase, and grey literature with search timeline extending up to January 2024. Studies that employed AI for 2D or 3D cephalometric landmark detection were included. The selection of studies, data extraction, and quality assessment of the included studies were performed independently by two reviewers. The risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. A meta-analysis was conducted to evaluate the accuracy of the 2D landmarks identification based on both mean radial error and standard error. Following the removal of duplicates, title and abstract screening, and full-text reading, 34 publications were selected. Amongst these, 27 studies evaluated the accuracy of AI-driven automated landmarking on 2D lateral cephalograms, while 7 studies involved 3D-CBCT images. A meta-analysis, based on the success detection rate of landmark placement on 2D images, revealed that the error was below the clinically acceptable threshold of 2 mm (1.39 mm; 95% confidence interval: 0.85–1.92 mm). For 3D images, meta-analysis could not be conducted due to significant heterogeneity amongst the study designs. However, qualitative synthesis indicated that the mean error of landmark detection on 3D images ranged from 1.0 to 5.8 mm. Both automated 2D and 3D landmarking proved to be time-efficient, taking less than 1 min. Most studies exhibited a high risk of bias in data selection (n = 27) and reference standard (n = 29). The performance of AI-driven cephalometric landmark detection on both 2D cephalograms and 3D-CBCT images showed potential in terms of accuracy and time efficiency. However, the generalizability and robustness of these AI systems could benefit from further improvement. PROSPERO: CRD42022328800.
4 November 2024 in Journal of Dentistry
To provide an overview of artificial intelligence (AI) applications in orthodontic diagnosis and treatment planning, and to evaluate whether AI improves accuracy, reliability, and time efficiency compared to expert-based manual approaches, while highlighting its current limitations. This review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist. An electronic search was performed on PubMed, Web of Science, and Embase electronic databases. Additional studies were identified from Google Scholar and by hand searching through included studies. The search was carried out until June 2023 without restriction of language and publication year. After applying the selection criteria, 71 articles were included in the review. The main research areas were classified into three domains based on the purpose of AI: diagnostics (n = 29), landmark identification (n = 20) and treatment planning (n = 22). This scoping review shows that AI can be used in various orthodontic diagnosis and treatment planning applications, with anatomical landmark detection being the most studied domain. While AI shows potential in improving time efficiency and reducing operator variability, the accuracy and reliability have not yet consistently surpassed those of expert clinicians. At all moments, human supervision remains essential. Further advances and optimizations are necessary to strive towards automated patient-specific treatment planning. AI in orthodontics has shown its ability to serve as a decision-support system, thereby enhancing the efficiency of diagnostics and treatment planning within orthodontics digital workflow."
15 August 2020 in Orthodontics & Craniofacial Research
31 July 2024 in Journal of Dentistry
This study aimed to validate a newly developed automated method (Virtual Patient Creator, Relu, Leuven, Belgium) for multimodal registration of intraoral scan (IOS) and Cone Beam Computed Tomography (CBCT). Time point-matched IOS and CBCT scans of forty patients with variable dental statuses (natural dentition, partial edentulism, presence of orthodontic brackets) were selected. Three operators registered IOS and CBCT scans using three state-of-the-art softwares for orthodontics and orthognathic surgery (IPS Case Designer, Proplan CMF and Dolphin Imaging). Automated registration was compared to expert-performed semi-automated registration. Time consumption, accuracy, and consistency of the proposed method were benchmarked to semi-automated registration using root mean squared error calculations. The robustness of the automated registration was evaluated in relationship to the dental status of the patients in the dataset. On average, automatic registration was 7.3 times faster than semi-automatic registration performed by an expert operator. Automatic registration yielded reliable results with low deviation errors compared to the differently skilled operators and semi-automated software. Automated registration surpassed human variability as expressed in intra- and inter-operator inconsistencies. Neither orthodontic brackets nor edentulism impacted registration accuracy. The presented automated method for IOS and CBCT registration is faster, equally accurate, and more consistent than semi-automatic registration performed by an expert or an occasional operator. With similar results among cases with different dental statuses, the clinical feasibility of the method is ensured. A validated automated registration method provides accurate and fast multimodal image integration without incorporating operator bias at the very start of the digital workflows for dentistry, periodontics, orthodontics and orthognathic surgery.
18 September 2025 in Journal of Dentistry
This scoping review aimed to identify evidence-based research papers on AI-enhanced treatment planning tools for orthodontics. The clinical relevance and applicability of academically validated AI tools were examined and complemented by an assessment of commercially available software. PubMed, Embase, Web of Science, and Cochrane were searched up to February 2025 for English-language studies on AI-based orthodontic treatment planning tools. Included studies were validation, accuracy, evaluation research papers; were available in full text; published in English; and focused on digital orthodontic treatment planning. Of 307 studies identified, 17 met inclusion criteria. The most focused on AI-driven decision-making for orthodontic extractions and design. Others explored automation for deep-bite planning and expansion in mixed dentition. Several studies evaluated large language models (LLMs) for answering orthodontic questions. Included were eight evaluation studies, three validation studies, two accuracy studies, and three comparative studies. AI methods included machine learning, deep learning, and LLMs, with reported accuracies from 72% to 95%. Seven commercial AI tools for orthodontic treatment planning were identified. The reviewed studies primarily addressed key treatment planning decisions or broader treatment recommendations. Academically validated tools typically rely on clinician-provided text inputs, whereas commercial AI solutions can process raw clinical data, such as intraoral scans. There is a mismatch between academically validated tools and commercially available systems, which generally lack published validation. This gap highlights the need for validation of commercial tools to ensure effective clinical integration. AI-based tools for orthodontic treatment planning might enhance clinical efficiency and promote consistency in decision-making. Both academic and commercial solutions demonstrate significant potential as decision-support systems, reinforcing - rather than replacing - the expertise of orthodontic professionals.