Mandible Morphometry Analysis Based on Stereolithography Measurements
13th International Conference on E-Health and Bioengineering, EHB 2025, Iasi, Romanya, 13 - 14 Kasım 2025, cilt.143 IFMBE, ss.245-251, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Cilt numarası: 143 IFMBE
- Doi Numarası: 10.1007/978-3-032-24045-3_27
- Basıldığı Şehir: Iasi
- Basıldığı Ülke: Romanya
- Sayfa Sayıları: ss.245-251
- Anahtar Kelimeler: CBCT, K-Means Clustering, Machine Learning, Mandible Morphometry, STL Models
- Ankara Üniversitesi Adresli: Evet
Özet
This study investigates the application of unsupervised machine learning for mandibular morphometric analysis using stereolithography (STL) models derived from cone-beam computed tomography (CBCT) images. AI-driven segmentation was employed to generate 3D STL representations of the mandible for 99 subjects (mean age = 27.3 years, SD = 4.2; 48 males, 51 females) with normal mandibular morphology. Five clinically relevant anatomical distances were extracted and analyzed using K-means clustering (K = 3). The analysis revealed distinct morphological subgroups, particularly pronounced in coronal plane measurements. These findings demonstrate the potential of clustering techniques in identifying structural patterns, offering insights for orthodontics, maxillofacial surgery, and forensic odontology.