Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

Transformer-Based Swin-UNETR for Multi-Structure Segmentation and Bone Measurement in CBCT-Based Dental Implant Planning

Journal of Robotics and Control (JRC) · Vol. 7 No. 1 (2026) · hlm. 3422–3432

DOI
10.18196/jrc.v7i1.26041
ISSN
2715-5072, 2715-5056
Terbit
2026-03-13
Penerbit
Universitas Muhammadiyah Yogyakarta
Bahasa
eng
Lisensi
creativecommons.org/licenses/by-sa/4.0

Abstrak

Dental implant planning requires precise tooth identification and alveolar bone measurement. Traditional assessment methods include manual interpretation of cone-beam computed tomography (CBCT) data, which is time-consuming and sensitive to observer variability. This paper presents an automated deep learning system for CBCT image segmentation and analysis of implant-related anatomical features to overcome these constraints. The transformer-based Swin-UNETR architecture captures both local spatial features and long-range contextual dependencies within volumetric CBCT data in volumetric dental data via hierarchical self-attention processes and an encoder–decoder structure. The model was trained and assessed using annotated CBCT datasets of missing teeth and bone structures. Dice Similarity Coefficient (DSC), loss convergence behavior, and training stability were used to evaluate performance. Experimental results show stable loss convergence and progressive segmentation improvement, achieving a Dice Similarity Coefficient (DSC) of 0.95, demonstrating accurate delineation of complex anatomical boundaries including the mandibular canal, maxillary sinus, and missing tooth regions and identifies clinically significant implant regions. The transformer-based architecture outperforms conventional 3D U-Net models in contextual representation and structural segmentation. The suggested approach reduces manual involvement and provides a reliable computational tool for implant site assessment. Transformer-driven deep learning models may improve automated dental image processing and computer-assisted implant planning systems

Kata kunci

Jaringan kolaborasi para penulis

4 penulis · 5 tautan kepenulisan bersama · 3 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Jain University
  • Nitte University
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Varuna KumaraJain University2

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://journal.umy.ac.id/index.php/jrc/oai, pertama di-harvest 9 Nov 2026, terakhir terlihat 9 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Transformer-Based Swin-UNETR for Multi-Structure Segmentation and Bone Measurement in CBCT-Based Dental Implant Planning</dc:title>
	<dc:creator>Nambiar, Rajashree</dc:creator>
	<dc:creator>Ranjith Bhat</dc:creator>
	<dc:creator>Raghu Nanjundegowda</dc:creator>
	<dc:subject xml:lang="en-US">Automated bone measurement</dc:subject>
	<dc:subject xml:lang="en-US">Cone Beam Computed Tomography (CBCT)</dc:subject>
	<dc:subject xml:lang="en-US">Deep Learning</dc:subject>
	<dc:subject xml:lang="en-US">Dental implant planning</dc:subject>
	<dc:subject xml:lang="en-US">Residual connection</dc:subject>
	<dc:subject xml:lang="en-US">Swin-UNETR segmentation</dc:subject>
	<dc:description xml:lang="en-US">Dental implant planning requires precise tooth identification and alveolar bone measurement. Traditional assessment methods include manual interpretation of cone-beam computed tomography (CBCT) data, which is time-consuming and sensitive to observer variability. This paper presents an automated deep learning system for CBCT image segmentation and analysis of implant-related anatomical features to overcome these constraints. The transformer-based Swin-UNETR architecture captures both local spatial features and long-range contextual dependencies within volumetric CBCT data in volumetric dental data via hierarchical self-attention processes and an encoder–decoder structure. The model was trained and assessed using annotated CBCT datasets of missing teeth and bone structures. Dice Similarity Coefficient (DSC), loss convergence behavior, and training stability were used to evaluate performance. Experimental results show stable loss convergence and progressive segmentation improvement, achieving a Dice Similarity Coefficient (DSC) of 0.95,  demonstrating accurate delineation of complex anatomical boundaries including the mandibular canal, maxillary sinus, and missing tooth regions and identifies clinically significant implant regions. The transformer-based architecture outperforms conventional 3D U-Net models in contextual representation and structural segmentation. The suggested approach reduces manual involvement and provides a reliable computational tool for implant site assessment. Transformer-driven deep learning models may improve automated dental image processing and computer-assisted implant planning systems</dc:description>
	<dc:publisher xml:lang="en-US">Universitas Muhammadiyah Yogyakarta</dc:publisher>
	<dc:date>2026-03-13</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.umy.ac.id/index.php/jrc/article/view/26041</dc:identifier>
	<dc:identifier>10.18196/jrc.v7i1.26041</dc:identifier>
	<dc:source xml:lang="en-US">Journal of Robotics and Control (JRC); Vol. 7 No. 1 (2026); 3422-3432</dc:source>
	<dc:source>2715-5072</dc:source>
	<dc:source>2715-5056</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.umy.ac.id/index.php/jrc/article/view/26041/12289</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2026 Rajashree Nambiar, Ranjith Bhat, Raghu Nanjundegowda</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>