Deep Learning-Based Anatomical Level Classification of Craniofacial CT Slices: Comparison of CNN Architectures and Ensemble Methods
- 1Hyogo Medical University
Abstract
Computed tomography (CT) interpretation requires recognition of the anatomical level of each slice. Classification may facilitate preprocessing and AI-assisted fracture diagnosis. We developed deep learning models to classify craniofacial CT slices into six anatomical categories. A total of 11,769 images were collected from a multicenter database involving 10 Japanese university hospitals and the public Head-Neck-Radiomics-HN1 dataset. Patients were divided into training (70%) and evaluation (30%) sets. Five CNN architectures were trained from scratch, and hard- and soft-voting ensembles were evaluated. Individual-model accuracy ranged from 94.72% to 96.09%; the hard-voting ensemble achieved 96.46% accuracy (Cohens {kappa} = 0.954). Although performance differed among individual models (Cochrans Q = 99.29, p < 0.001), the ensemble did not significantly outperform EfficientNetV2B1 (McNemars p = 0.163). Errors occurred mainly between adjacent anatomical regions. Deep learning can classify craniofacial CT slices accurately and may support automated dataset organization and AI-assisted fracture diagnosis. External validation is needed.
Related
- Anatomical comparison populations and threshold transport in vascular transcriptomic classification
- Multi-Model Comparison of Deep and Classical Architectures for ECG Diagnostic Superclass Classification on PTB-XL
- CT Radiomics and Machine Learning for Three-Class Bone Density Classification: An Exploratory Comparison Across Lumbar Vertebral Levels
- Anatomical spread of the novel transversus abdominis–rectus abdominis aponeurotic (TARAA) block: a cadaveric comparison with lateral transversus abdominis plane (TAP) and combined TAP-rectus sheath blocks
- Artificial Intelligence and Anatomical Variation: A Narrative Review of Detection, Classification, and Educational Applications.
BibTeX
@misc{nishimoto2026deep,
title = {{Deep Learning-Based Anatomical Level Classification of Craniofacial CT Slices: Comparison of CNN Architectures and Ensemble Methods}},
author = {S. Nishimoto and K. Nakajima and H. Yoshiike and H. Ishise and T. Saito and K. Kawai},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.16.26365710},
url = {https://www.medrxiv.org/content/10.64898/2026.10.16.26365710v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Deep Learning-Based Anatomical Level Classification of Craniofacial CT Slices: Comparison of CNN Architectures and Ensemble Methods
AU - S. Nishimoto
AU - K. Nakajima
AU - H. Yoshiike
AU - H. Ishise
AU - T. Saito
AU - K. Kawai
PY - 2026
DA - 2026/10/16
PB - medRxiv
DO - 10.64898/2026.10.16.26365710
UR - https://www.medrxiv.org/content/10.64898/2026.10.16.26365710v1
AB - Computed tomography (CT) interpretation requires recognition of the anatomical level of each slice. Classification may facilitate preprocessing and AI-assisted fracture diagnosis. We developed deep learning models to classify craniofacial CT slices into six anatomical categories. A total of 11,769 images were collected from a multicenter database involving 10 Japanese university hospitals and the public Head-Neck-Radiomics-HN1 dataset. Patients were divided into training (70%) and evaluation (30%) sets. Five CNN architectures were trained from scratch, and hard- and soft-voting ensembles were evaluated. Individual-model accuracy ranged from 94.72% to 96.09%; the hard-voting ensemble achieved 96.46% accuracy (Cohens {kappa} = 0.954). Although performance differed among individual models (Cochrans Q = 99.29, p < 0.001), the ensemble did not significantly outperform EfficientNetV2B1 (McNemars p = 0.163). Errors occurred mainly between adjacent anatomical regions. Deep learning can classify craniofacial CT slices accurately and may support automated dataset organization and AI-assisted fracture diagnosis. External validation is needed.
ER -