[Exploration and clinical application of the "digital and intelligent surgery" diagnosis and treatment workflow for oral and maxillofacial tumors].

内容摘要

Tumors in the oral and maxillofacial region present significant clinical challenges due to anatomical complexity and high individual variability, with the traditional experience-dependent model often lacking three-dimensional visualization, precise intraoperative navigation, and quantitative postoperative assessment. This article comprehensively reviews over a decade of research and clinical advances in "digital and intelligent surgery" developed by our team at Peking University School and Hospital of Stomatology, systematically documenting its transformative impact on tumor management. In digital surgery, we have established multimodal image fusion techniques integrating CT, MRI, and PET/CT to achieve detailed three-dimensional preoperative visualization, enabling accurate delineation of tumor boundaries and relationships with critical anatomical structures, such as nerves and vessels. We further developed personalized surgical planning methods including virtual design for jaw reconstruction using vascularized fibula or iliac crest flaps, computer-aided pre-forming of orbital titanium mesh, 3D-printed patient- specific plates manufactured via electron beam melting, soft-tissue flap simulation and volumetric planning for the anterolateral thigh flap, and implant-guided rehabilitation for complex maxillary defects. For surgical execution, navigation systems and mixed reality technologies have been implemented to enable accurate tumor resection, osteotomy guidance, and precise positioning of reconstructed bone segments, thereby enhancing surgical accuracy and safety while reducing operative time. In parallel, artificial intelligence has been integrated to enhance diagnostic and planning efficiency through deep learning-based tumor segmentation and classification from enhanced CT and MRI, automated reconstruction planning based on shape completion and morphometric descriptors, postoperative facial contour prediction using surface mesh deformation models, and machine learning-driven prognostic modeling for salivary gland malignancies based on clinicopathological data. The synergistic integration of these digital and intelligent technologies, collectively termed "digital and intelligent surgery", has shifted clinical practice from an experience-driven to a data-driven paradigm, significantly improving precision, safety, and efficiency while enabling truly personalized treatment pathways. This review also identifies current limitations such as the need for further automation in soft-tissue simulation and broader clinical validation of AI tools, and outlines future directions including the development of integrated surgical platforms and real-time adaptive planning systems, emphasizing the role of intelligent surgical systems in shaping the next generation of oral and maxillofacial oncology care toward more predictive, preventive, and patient-centered outcomes. 发生于口腔颌面部的肿瘤因解剖结构复杂且个体差异大, 传统"经验依赖型"诊疗模式存在术前规划无法三维可视化、术中缺少精准导航、术后缺乏量化评估等局限性。本文系统综述了本课题组十余年来在口腔颌面部肿瘤"数智化外科"领域的探索与临床应用成果。在数字化外科方面, 课题组建立了基于CT、MRI、PET/CT等多模态数据融合的术前三维可视化技术, 开发了颌骨缺损重建、眶底钛网预成形、3D打印个性化钛板、软组织皮瓣虚拟设计与种植修复等个性化手术方案设计方法, 并将外科导航系统、混合现实技术应用于术中精准定位与肿瘤切除。在人工智能应用方面, 课题组探索了基于深度学习的肿瘤影像自动分割与分类、颌骨重建方案自动生成、术后面型预测及唾液腺恶性肿瘤预后评估等智能化技术。通过数字化与智能化技术的深度融合, "数智化外科"实现了从经验驱动到数据驱动的诊疗模式转型, 显著提升了口腔颌面部肿瘤诊疗的精准性、安全性和效率, 为患者提供了更加个性化、可预测的治疗方案。本课题组还展望了未来"数智化外科"在口腔颌面部肿瘤诊疗中的发展方向。