Scoping Review of Automatic Landmark Detection in 3D CT/CBCT Imaging

A scoping review synthesizing 37 studies on automatic landmark identification in three-dimensional computed tomography and cone-beam computed tomography. The review, authored by You Wu and published on figshare in 2026, focuses on AI-based methods and explores progress in accuracy, efficiency, and clinical applicability. It includes studies published until January 2026, sourced from PubMed, Web of Science, and the Cochrane Library.

Use Cases

• Benchmarking AI-based landmark detection algorithms based on the reported accuracy metrics and clinical correctness standards.
• Identifying research gaps in automatic landmarking for soft-tissue structures based on the review's finding that only 6 of 37 studies involved soft-tissue landmarks.
• Assessing the generalizability of single algorithms for clinical use based on the review's conclusions about limited robustness.
• Evaluating the standardization of accuracy metrics in 3D space based on the review's discussion of non-standardized evaluation systems.

Strengths

• Synthesizes findings from 37 distinct studies, providing a consolidated view of the field.
• Explicitly states the search cut-off date (January 2026) and the databases used (PubMed, Web of Science, Cochrane Library).
• Provides quantitative analysis, such as 56.8% of studies met the clinical correctness standard of MRE < 2mm.

Limitations

• The dataset is a 36.1 KB DOCX document, indicating a limited scope focused on review text rather than primary data.
• Row and column counts are unknown, as the content is a review document, not a structured data table.
• The description does not specify the exact features or data points extracted from each reviewed study for analysis.

Provenance

Source

figshare

Collection Method

Scoping review of literature from PubMed, Web of Science, and the Cochrane Library.

Time Range

Studies published until January 2026.

Freshness

Last updated 2026-05-29 06:07:36.