Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network
<b>Background</b>: Accurate determination of the natural head position (NHP) is essential in orthognathic surgery for optimal surgical planning and improved patient outcomes. However, traditional methods encounter reproducibility issues and rely on external devices or patient cooperation...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/1/42 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549269809496064 |
|---|---|
| author | Ji-Yong Yoo Su Yang Sang-Heon Lim Ji Yong Han Jun-Min Kim Jo-Eun Kim Kyung-Hoe Huh Sam-Sun Lee Min-Suk Heo Hoon Joo Yang Won-Jin Yi |
| author_facet | Ji-Yong Yoo Su Yang Sang-Heon Lim Ji Yong Han Jun-Min Kim Jo-Eun Kim Kyung-Hoe Huh Sam-Sun Lee Min-Suk Heo Hoon Joo Yang Won-Jin Yi |
| author_sort | Ji-Yong Yoo |
| collection | DOAJ |
| description | <b>Background</b>: Accurate determination of the natural head position (NHP) is essential in orthognathic surgery for optimal surgical planning and improved patient outcomes. However, traditional methods encounter reproducibility issues and rely on external devices or patient cooperation, potentially leading to inaccuracies in the surgical plan. <b>Methods</b>: To address these limitations, we developed a geometric deep learning network (NHP-Net) to automatically reproduce NHP from CT scans. A dataset of 150 orthognathic surgery patients was utilized. Three-dimensional skull meshes were converted into point clouds and normalized to fit within a unit sphere. NHP-Net was trained to predict a 3 × 3 rotation matrix to align the CT-acquired posture with the NHP. Experiments were conducted to determine optimal point cloud sizes and loss functions. Performance was evaluated using mean absolute error (MAE) for roll, pitch, and yaw angles, as well as a rotation error (RE) metric. <b>Results</b>: NHP-Net achieved the lowest RE of 1.918° ± 1.099° and demonstrated significantly lower MAEs in roll and pitch angles compared to other deep learning models (<i>p</i> < 0.05). These findings indicate that NHP-Net can accurately align CT-acquired postures to the NHP, enhancing the precision of surgical planning. <b>Conclusions</b>: By effectively improving the accuracy and efficiency of NHP reproduction, NHP-Net reduces the workload of surgeons, supports more precise orthognathic surgical interventions, and ultimately contributes to better patient outcomes. |
| format | Article |
| id | doaj-art-5c3966fe0c364fe79c9fb703915f280b |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-5c3966fe0c364fe79c9fb703915f280b2025-01-10T13:16:32ZengMDPI AGDiagnostics2075-44182024-12-011514210.3390/diagnostics15010042Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning NetworkJi-Yong Yoo0Su Yang1Sang-Heon Lim2Ji Yong Han3Jun-Min Kim4Jo-Eun Kim5Kyung-Hoe Huh6Sam-Sun Lee7Min-Suk Heo8Hoon Joo Yang9Won-Jin Yi10Department of Biomedical Radiation Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Seoul 08826, Republic of KoreaDepartment of Applied Bioengineering, Graduate School of Convergence Science and Technology, Seoul National University, Seoul 08826, Republic of KoreaInterdisciplinary Program in Bioengineering, Graduate School of Engineering, Seoul National University, Seoul 08826, Republic of KoreaInterdisciplinary Program in Bioengineering, Graduate School of Engineering, Seoul National University, Seoul 08826, Republic of KoreaDepartment of Electronics and Information Engineering, Hansung University, Seoul 02876, Republic of KoreaDepartment of Oral and Maxillofacial Radiology, Dental Research Institute, School of Dentistry, Seoul National University, Seoul 03080, Republic of KoreaDepartment of Oral and Maxillofacial Radiology, Dental Research Institute, School of Dentistry, Seoul National University, Seoul 03080, Republic of KoreaDepartment of Oral and Maxillofacial Radiology, Dental Research Institute, School of Dentistry, Seoul National University, Seoul 03080, Republic of KoreaDepartment of Oral and Maxillofacial Radiology, Dental Research Institute, School of Dentistry, Seoul National University, Seoul 03080, Republic of KoreaDepartment of Oral and Maxillofacial Surgery, Dental Research Institute, School of Dentistry, Seoul National University, Seoul 03080, Republic of KoreaDepartment of Applied Bioengineering, Graduate School of Convergence Science and Technology, Seoul National University, Seoul 08826, Republic of Korea<b>Background</b>: Accurate determination of the natural head position (NHP) is essential in orthognathic surgery for optimal surgical planning and improved patient outcomes. However, traditional methods encounter reproducibility issues and rely on external devices or patient cooperation, potentially leading to inaccuracies in the surgical plan. <b>Methods</b>: To address these limitations, we developed a geometric deep learning network (NHP-Net) to automatically reproduce NHP from CT scans. A dataset of 150 orthognathic surgery patients was utilized. Three-dimensional skull meshes were converted into point clouds and normalized to fit within a unit sphere. NHP-Net was trained to predict a 3 × 3 rotation matrix to align the CT-acquired posture with the NHP. Experiments were conducted to determine optimal point cloud sizes and loss functions. Performance was evaluated using mean absolute error (MAE) for roll, pitch, and yaw angles, as well as a rotation error (RE) metric. <b>Results</b>: NHP-Net achieved the lowest RE of 1.918° ± 1.099° and demonstrated significantly lower MAEs in roll and pitch angles compared to other deep learning models (<i>p</i> < 0.05). These findings indicate that NHP-Net can accurately align CT-acquired postures to the NHP, enhancing the precision of surgical planning. <b>Conclusions</b>: By effectively improving the accuracy and efficiency of NHP reproduction, NHP-Net reduces the workload of surgeons, supports more precise orthognathic surgical interventions, and ultimately contributes to better patient outcomes.https://www.mdpi.com/2075-4418/15/1/42orthognathic surgerycomputed tomographynatural head positionhead pose estimationgeometric deep learning |
| spellingShingle | Ji-Yong Yoo Su Yang Sang-Heon Lim Ji Yong Han Jun-Min Kim Jo-Eun Kim Kyung-Hoe Huh Sam-Sun Lee Min-Suk Heo Hoon Joo Yang Won-Jin Yi Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network Diagnostics orthognathic surgery computed tomography natural head position head pose estimation geometric deep learning |
| title | Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network |
| title_full | Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network |
| title_fullStr | Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network |
| title_full_unstemmed | Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network |
| title_short | Automatic Reproduction of Natural Head Position in Orthognathic Surgery Using a Geometric Deep Learning Network |
| title_sort | automatic reproduction of natural head position in orthognathic surgery using a geometric deep learning network |
| topic | orthognathic surgery computed tomography natural head position head pose estimation geometric deep learning |
| url | https://www.mdpi.com/2075-4418/15/1/42 |
| work_keys_str_mv | AT jiyongyoo automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT suyang automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT sangheonlim automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT jiyonghan automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT junminkim automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT joeunkim automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT kyunghoehuh automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT samsunlee automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT minsukheo automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT hoonjooyang automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork AT wonjinyi automaticreproductionofnaturalheadpositioninorthognathicsurgeryusingageometricdeeplearningnetwork |