Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization
Mastoidectomy is critical in acoustic neuroma surgery, where precise planning of the bone milling area is essential for surgical navigation. The complexity of representing the irregular volumetric area and the presence of high-risk structures (e.g., blood vessels and nerves) complicate this task. In...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/448 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587542668509184 |
|---|---|
| author | Sheng Yang Haowei Li Peihai Zhang Wenqing Yan Zhe Zhao Hui Ding Guangzhi Wang |
| author_facet | Sheng Yang Haowei Li Peihai Zhang Wenqing Yan Zhe Zhao Hui Ding Guangzhi Wang |
| author_sort | Sheng Yang |
| collection | DOAJ |
| description | Mastoidectomy is critical in acoustic neuroma surgery, where precise planning of the bone milling area is essential for surgical navigation. The complexity of representing the irregular volumetric area and the presence of high-risk structures (e.g., blood vessels and nerves) complicate this task. In order to determine the bone area to mill using preoperative CT images automatically, we propose an automated planning method using evolutionary multi-objective optimization for safer and more efficient milling plans. High-resolution segmentation of the adjacent risk structures is performed on preoperative CT images with a template-based approach. The maximum milling area is defined based on constraints from the risk structures and tool dimensions. Deformation fields are used to simplify the volumetric area into limited continuous parameters suitable for optimization. Finally, a multi-objective optimization algorithm is used to achieve a Pareto-optimal design. Compared with manual planning on six volumes, our method reduced the potential damage to the scala vestibuli by 29.8%, improved the milling boundary smoothness by 78.3%, and increased target accessibility by 26.4%. Assessment by surgeons confirmed the clinical feasibility of the generated plans. In summary, this study presents a parameterization approach to irregular volumetric regions, enabling automated milling area planning through optimization techniques that ensure safety and feasibility. This method is also adaptable to various volumetric planning scenarios. |
| format | Article |
| id | doaj-art-fe98c62f84f349aaa1cf7c3a2e20ab9a |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-fe98c62f84f349aaa1cf7c3a2e20ab9a2025-01-24T13:48:58ZengMDPI AGSensors1424-82202025-01-0125244810.3390/s25020448Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective OptimizationSheng Yang0Haowei Li1Peihai Zhang2Wenqing Yan3Zhe Zhao4Hui Ding5Guangzhi Wang6School of Biomedical Engineering, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaSchool of Biomedical Engineering, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaDepartment of Neurosurgery, Beijing Tsinghua Changgung Hospital, Li Tang Road, Beijing 100043, ChinaSchool of Biomedical Engineering, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaSchool of Clinical Medicine, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaSchool of Biomedical Engineering, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaSchool of Biomedical Engineering, Tsinghua University, Shuang Qing Road, Beijing 100084, ChinaMastoidectomy is critical in acoustic neuroma surgery, where precise planning of the bone milling area is essential for surgical navigation. The complexity of representing the irregular volumetric area and the presence of high-risk structures (e.g., blood vessels and nerves) complicate this task. In order to determine the bone area to mill using preoperative CT images automatically, we propose an automated planning method using evolutionary multi-objective optimization for safer and more efficient milling plans. High-resolution segmentation of the adjacent risk structures is performed on preoperative CT images with a template-based approach. The maximum milling area is defined based on constraints from the risk structures and tool dimensions. Deformation fields are used to simplify the volumetric area into limited continuous parameters suitable for optimization. Finally, a multi-objective optimization algorithm is used to achieve a Pareto-optimal design. Compared with manual planning on six volumes, our method reduced the potential damage to the scala vestibuli by 29.8%, improved the milling boundary smoothness by 78.3%, and increased target accessibility by 26.4%. Assessment by surgeons confirmed the clinical feasibility of the generated plans. In summary, this study presents a parameterization approach to irregular volumetric regions, enabling automated milling area planning through optimization techniques that ensure safety and feasibility. This method is also adaptable to various volumetric planning scenarios.https://www.mdpi.com/1424-8220/25/2/448volumetric surgical planningvolumetric area parameterizationmulti-objective optimizationbone milling |
| spellingShingle | Sheng Yang Haowei Li Peihai Zhang Wenqing Yan Zhe Zhao Hui Ding Guangzhi Wang Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization Sensors volumetric surgical planning volumetric area parameterization multi-objective optimization bone milling |
| title | Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization |
| title_full | Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization |
| title_fullStr | Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization |
| title_full_unstemmed | Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization |
| title_short | Automated Volumetric Milling Area Planning for Acoustic Neuroma Surgery via Evolutionary Multi-Objective Optimization |
| title_sort | automated volumetric milling area planning for acoustic neuroma surgery via evolutionary multi objective optimization |
| topic | volumetric surgical planning volumetric area parameterization multi-objective optimization bone milling |
| url | https://www.mdpi.com/1424-8220/25/2/448 |
| work_keys_str_mv | AT shengyang automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT haoweili automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT peihaizhang automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT wenqingyan automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT zhezhao automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT huiding automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization AT guangzhiwang automatedvolumetricmillingareaplanningforacousticneuromasurgeryviaevolutionarymultiobjectiveoptimization |