{
  "abstract": "Objective: This study aimed to develop and validate a smartphone-based pipeline for generating clinically interpretable three-dimensional (3D) facial models using locally deployable and low-cost hardware. Methods: A five-phase reconstruction pipeline was developed, including image acquisition, foreground matting, camera pose estimation, two-dimensional Gaussian Splatting (2DGS) reconstruction, and metric scale calibration. Reconstruction accuracy was evaluated against a reference 3dMD stereophotogrammetry system using Hausdorff distance analysis, linear regression, intraclass correlation coefficients (ICC), and Bland–Altman analysis of five clinically relevant facial landmark measurements. Results: The reconstructed models generated by the proposed pipeline demonstrated good geometric agreement with the reference 3dMD models. Mean Hausdorff and root mean square (RMS) errors remained low across all participants. Landmark-based analyses demonstrated strong agreement between the reconstructed and reference models, with high ICC values, strong linear correlations, minimal systematic bias, and clinically acceptable limits of agreement. Absolute measurement errors remained within clinically meaningful ranges. Conclusion: The proposed pipeline enables accurate metric 3D facial reconstruction using widely available mobile devices and consumer-grade computing hardware. It demonstrated clinically acceptable agreement with a reference stereophotogrammetry system and may provide a practical alternative for facial assessment, treatment planning, and longitudinal follow-up in orthodontic and craniofacial applications. Clinical significance: The proposed pipeline enables the generation of clinically measurable 3D craniofacial models with accuracy comparable to commercial stereophotogrammetry systems, offering a low-cost and scalable approach that may facilitate the adoption of quantitative craniofacial assessment in routine dental settings.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Orthodontics, Jinan Stomatological Hospital, Shandong Provincial Health Commission Key Laboratory of Oral Diseases and Tissue Regeneration, Shandong Medical and Pharmaceutical University, Jinan, Shandong, 250000, China; Corresponding author. No. 101, Jingliu Road, Jinan, Shandong, China.",
    "GreenOne Technology Consulting Co., Ltd., Jinan, Shandong, 250000, China",
    "Department of Orthodontics, Jinan Stomatological Hospital, Shandong Provincial Health Commission Key Laboratory of Oral Diseases and Tissue Regeneration, Shandong Medical and Pharmaceutical University, Jinan, Shandong, 250000, China",
    "Department of Orthodontics, Jinan Stomatological Hospital, Jinan, Shandong, 250000, China"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Fengcong Wang"
    },
    {
      "affs": [
        1
      ],
      "name": "Yang Lyu"
    },
    {
      "affs": [
        2
      ],
      "name": "Ping Liu"
    },
    {
      "affs": [
        3
      ],
      "name": "Donglai Liu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/17mfwfysjcs5t.bib",
  "cited_by": 0,
  "doi": "10.1016/j.ddj.2026.100125",
  "fulltext_url": "http://www.sciencedirect.com/science/article/pii/S2950643326000590",
  "id": "17mfwfysjcs5t",
  "issns": [
    "2950-6433"
  ],
  "landing_url": "https://doi.org/10.1016/j.ddj.2026.100125",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "lcc",
      "R855-855.5"
    ],
    [
      "lcc",
      "RK1-715"
    ],
    [
      "openalex",
      "3505|Orthodontics and Dentofacial Orthopedics"
    ]
  ],
  "openalex": "W7214799413",
  "pdf_url": "https://www.sciencedirect.com/science/article/pii/S2950643326000590/pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-30",
  "publisher": "Elsevier",
  "ris": "https://research.draconitic.com/paper/17mfwfysjcs5t.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T13:44:32Z",
      "native_id": "70ecf8811a16421b993ddfde8fed7458",
      "source": "doaj",
      "url": "https://doi.org/10.1016/j.ddj.2026.100125"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "lcc:R855-855.5",
      "subfield": "general"
    }
  ],
  "title": "Bringing high-fidelity 3D craniofacial assessment to the chairside: A low-cost smartphone pipeline via 2D Gaussian splatting",
  "title_text": "Bringing high-fidelity 3D craniofacial assessment to the chairside: A low-cost smartphone pipeline via 2D Gaussian splatting",
  "type": "peer-reviewed",
  "updated_at": "2026-10-30T13:44:32Z",
  "url": "https://research.draconitic.com/paper/17mfwfysjcs5t",
  "venue": "Digital Dentistry Journal",
  "versions": []
}