Non-Cochrane · DTA · PM42698514
Parameter-level comparison of artificial intelligence and manual cephalometric measurements: a systematic review and meta-analysis.
Coronel-Zubiate Franz Tito, Marroquín-Soto Consuelo, Meza-Málaga Joan Manuel, et al. · Frontiers in dental medicine · 2026 · PubMed · doi:10.3389/fdmed.2026.1820984
Findings
Population Patients requiring orthodontic diagnosis
16 findings
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | L1–NB distance (mm) | SMD (Hedges’ g) 0.29 [-0.08, 0.63] GRADE not reported 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | L1–NB angle (°) | SMD (Hedges’ g) -0.22 [-0.96, 0.53] GRADE not reported 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | Interincisal angle (°) | SMD (Hedges’ g) 0.19 [-0.04, 0.42] GRADE not reported 3 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | Wits appraisal (mm) | SMD (Hedges’ g) 0.14 [-0.1, 0.38] GRADE not reported 3 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | ANB angle (°) | SMD (Hedges’ g) 0.28 [0.03, 0.53] Low 6 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | FMA angle (°) | SMD (Hedges’ g) -0.47 [-0.99, 0.05] Low Downgraded: imprecision, inconsistency 4 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | L1–MeGo angle (°) | SMD (Hedges’ g) -0.12 [-0.4, 0.15] GRADE not reported 4 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | IMPA (°) | SMD (Hedges’ g) -0.03 [-0.28, 0.22] Moderate 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | SNA angle (°) | SMD (Hedges’ g) 0.11 [-0.03, 0.24] Moderate 6 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | Gonial angle (°) | SMD (Hedges’ g) -0.59 [-1.9, 0.73] GRADE not reported 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | SNB angle (°) | SMD (Hedges’ g) -0.12 [-0.28, 0.04] Moderate 6 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | SN–MP angle (°) | SMD (Hedges’ g) -0.03 [-0.34, 0.28] GRADE not reported 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | U1–NA angle (°) | SMD (Hedges’ g) -0.11 [-0.5, 0.39] GRADE not reported 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | U1–SN angle (°) | SMD (Hedges’ g) -0.18 [-0.38, 0.01] GRADE not reported 4 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | U1–NA distance (mm) Unresolved: the two extractors disagree on this row. | SMD (Hedges’ g) -0.29 [-1.01, 0.44] Low 2 studies |
| AI-based cephalometric measurements | Manual cephalometric tracings by orthodontists or expert reference standards | SUM–MAN angle (°) Unresolved: the two extractors disagree on this row. | SMD (Hedges’ g) 0.55 [-0.52, 1.61] Low Downgraded: imprecision, inconsistency 3 studies |
Included primary studies 37
Open in Studies| Study | Year | Also included by | Outcomes it informs here | Trial data |
|---|---|---|---|---|
| Bao 2023 Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence DOI:10.1186/s12903-023-02881-8 | 2023 | this review only | not stated | not extracted yet |
| Bao 2023 Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence. PMID:37005593 | 2023 | this review only | not stated | not extracted yet |
| Blum 2023 Evaluation of an artificial intelligence–based algorithm for automated localization of craniofacial landmarks DOI:10.1007/s00784-023-04978-4 | 2023 | this review only | not stated | not extracted yet |
| Blum 2023 Evaluation of an artificial intelligence-based algorithm for automated localization of craniofacial landmarks. PMID:37014502 | 2023 | this review only | not stated | not extracted yet |
| Bor 2024 Comparison of AI-assisted cephalometric analysis and orthodontist-performed digital tracing analysis DOI:10.1186/s40510-024-00539-x | 2024 | this review only | not stated | not extracted yet |
| Bor 2024 Comparison of AI-assisted cephalometric analysis and orthodontist-performed digital tracing analysis. PMID:39428414 | 2024 | this review only | not stated | not extracted yet |
| Bulatova 2021 Assessment of automatic cephalometric landmark identification using artificial intelligence DOI:10.1111/ocr.12542 | 2021 | this review only | not stated | not extracted yet |
| Bulatova 2021 Assessment of automatic cephalometric landmark identification using artificial intelligence. PMID:34842346 | 2021 | this review only | not stated | not extracted yet |
| Jeon 2021 Comparison of cephalometric measurements between conventional and automatic cephalometric analysis using convolutional neural network DOI:10.1186/s40510-021-00358-4 | 2021 | this review only | not stated | not extracted yet |
| Jeon 2021 Comparison of cephalometric measurements between conventional and automatic cephalometric analysis using convolutional neural network. PMID:34056670 | 2021 | this review only | not stated | not extracted yet |
| Katyal 2022 Evaluation of the accuracy and reliability of WebCeph – An artificial intelligence-based online software DOI:10.25259/apos_138_2021 | 2022 | this review only | not stated | not extracted yet |
| Katyal 2022 matched by author and year (probable) | NR | this review only | not stated | not extracted yet |
| Khabadze 2024 Comparative Analysis of 3D Cephalometry Provided with Artificial Intelligence and Manual Tracing. PMID:39594190 | 2024 | this review only | not stated | not extracted yet |
| Kim KH 2025 Evaluation of the clinical utility of lateral cephalometry reconstructed from computed tomography extracted by artificial intelligence DOI:10.1016/j.jcms.2024.12.004 | 2025 | this review only | not stated | not extracted yet |
| Kim KH 2025 Evaluation of the clinical utility of lateral cephalometry reconstructed from computed tomography extracted by artificial intelligence. PMID:39658471 | 2025 | this review only | not stated | not extracted yet |
| Kim YH 2021 Influence of the Depth of the Convolutional Neural Networks on an Artificial Intelligence Model for Diagnosis of Orthognathic Surgery. PMID:33946874 | 2021 | this review only | not stated | not extracted yet |
| Koz 2025 Artificial Intelligence-Supported and App-Aided Cephalometric Analysis: Which One Can We Trust? PMID:40075806 | 2025 | this review only | not stated | not extracted yet |
| Kunz 2025 Assessment of the quality of different commercial providers using artificial intelligence for automated cephalometric analysis compared to… DOI:10.1007/s00056-023-00491-1 | 2023 | this review only | not stated | not extracted yet |
| Kunz 2025 Assessment of the quality of different commercial providers using artificial intelligence for automated cephalometric analysis compared to… PMID:37642657 | 2025 | this review only | not stated | not extracted yet |
| Lee 2023 Commercial artificial intelligence lateral cephalometric analysis: part 2—effects of human examiners on artificial intelligence… DOI:10.22514/jocpd.2023.087 | 2023 | this review only | not stated | not extracted yet |
| Lee 2023 Commercial artificial intelligence lateral cephalometric analysis: part 2-effects of human examiners on artificial intelligence… PMID:37997244 | 2023 | this review only | not stated | not extracted yet |
| Mahto 2022 Evaluation of fully automated cephalometric measurements obtained from web-based artificial intelligence driven platform DOI:10.1186/s12903-022-02170-w | 2022 | this review only | not stated | not extracted yet |
| Mahto 2022 Evaluation of fully automated cephalometric measurements obtained from web-based artificial intelligence driven platform. PMID:35440037 | 2022 | this review only | not stated | not extracted yet |
| Mehra 2024 VGG-16 based Deep Learning Approach for Cephalometric Landmark Detection DOI:10.2174/0118749445338097240923063441 | 2024 | this review only | not stated | not extracted yet |
| Menezes 2023 Assessment of landmark detection in cephalometric radiographs with different conditions of brightness and contrast using the an artificial… DOI:10.1259/dmfr.20230065 | 2023 | this review only | not stated | not extracted yet |
| Menezes 2023 Assessment of landmark detection in cephalometric radiographs with different conditions of brightness and contrast using the an artificial… PMID:37869886 | 2023 | this review only | not stated | not extracted yet |
| Mercier 2024 Reliability and accuracy of Artificial intelligence-based software for cephalometric diagnosis. A diagnostic study DOI:10.1186/s12903-024-05097-6 | 2024 | this review only | not stated | not extracted yet |
| Mercier 2024 Reliability and accuracy of Artificial intelligence-based software for cephalometric diagnosis. A diagnostic study. PMID:39468520 | 2024 | this review only | not stated | not extracted yet |
| O'Friel 2024 Assessing AudaxCeph®'s cephalometric tracing technology versus a semi-automated approach for analyzing severe Class II and Class III… PMID:39378572 | 2024 | this review only | not stated | not extracted yet |
| O’Friel 2024 Assessing AudaxCeph®’s cephalometric tracing technology versus a semi-automated approach for analyzing severe Class II and Class III… DOI:10.1016/j.ortho.2024.100926 | 2024 | this review only | not stated | not extracted yet |
| Panesar 2023 Precision and Accuracy Assessment of Cephalometric Analyses Performed by Deep Learning Artificial Intelligence with and without Human… DOI:10.3390/app13126921 | 2023 | this review only | not stated | not extracted yet |
| Park 2024 Clinical validity and precision of deep learning-based cone-beam computed tomography automatic landmarking algorithm. PMID:39371307 | 2024 | this review only | not stated | not extracted yet |
| Ramadan 2022 Convolution neural network based automatic localization of landmarks on lateral x-ray images DOI:10.1007/s11042-021-11596-3 | 2022 | this review only | not stated | not extracted yet |
| Ye 2023 Is automatic cephalometric software using artificial intelligence better than orthodontist experts in landmark identification? DOI:10.1186/s12903-023-03188-4 | 2023 | this review only | not stated | not extracted yet |
| Ye 2023 Is automatic cephalometric software using artificial intelligence better than orthodontist experts in landmark identification? PMID:37422630 | 2023 | this review only | not stated | not extracted yet |
| Zhang 2024 The concept of AI-assisted self-monitoring for skeletal malocclusion DOI:10.1177/14604582241274511 | 2024 | this review only | not stated | not extracted yet |
| Zhang 2024 The concept of AI-assisted self-monitoring for skeletal malocclusion. PMID:39120929 | 2024 | this review only | not stated | not extracted yet |
Review question
| Element | Planned (methods) | Found (results) |
|---|---|---|
| Population | Patients (children, adolescents or adults) requiring orthodontic diagnosis | Patients undergoing cephalometric analysis (n=22 studies; sample sizes not summarized in the main text) |
| Intervention | AI-based diagnostic methods for cephalometric analysis (machine learning and deep learning models) | Commercial and bespoke AI systems (WebCeph, CephX, DentaliQ.ortho, AudaxCeph, RadioCef, NemoCeph; CNN and conventional machine learning) (n=22) |
| Comparison | Manual cephalometric tracings performed by orthodontists or expert reference standards | Manual tracing by expert orthodontists or experienced residents (n=22) |
| Outcomes | Primary: parameter-level cephalometric measurements (e.g., ANB, SNA, SNB, IMPA, FMA, U1-NA, L1-NB) pooled as Hedges' g; secondary, narrative only: diagnostic metrics (sensitivity, specificity, AUC, accuracy, kappa/ICC) | 16 pooled parameter-level cephalometric measurements (U1-NA angle and mm, L1-NB angle and mm, IMPA, U1-SN, interincisal angle, L1-MeGo, SNA, SNB, FMA, SN-MP, ANB, Wits, gonial angle, SUM-MAN); diagnostic metrics summarized narratively |
| Study designs | Analytical observational studies, including comparative cross-sectional and diagnostic validation studies | Mostly single-centre observational comparisons on 2D lateral cephalograms or CBCT (n=22) |
| Timing | Not specified | Not specified |