Current and Future Central Nervous System Surgery-Part 1: Cervical Spinal Cord Compression.

From static to robotic: evolving navigation systems in oral and maxillofacial surgery.

Background: Achieving adequate margins in pelvic bone tumor resection remains difficult, as conventional navigation provides no direct three-dimensional margin feedback. We proposed an innovative dangerous region generation method based on 3D image resampling and anisotropic distance transform, integrated with computer-assisted navigation, to enhance surgical margin accuracy. This study aimed to evaluate its oncological safety, functional outcomes, and perioperative efficacy in pelvic tumor surgery. Methods: The study was conducted on 19 patients (8 males, 11 females) with primary pelvic bone tumors between May 2018 and June 2024. The age range was 19 to 66 years (mean age: 62.67 years). Histological diagnoses included chondrosarcoma (n = 6), giant cell tumor (n = 4), osteosarcoma (n = 1), chordoma (n = 2), Ewing sarcoma (n = 3), spindle cell sarcoma (n = 1), chondromyxoid fibroma (n = 1), and peripheral nerve sheath tumor (n = 1). The feasibility of the dangerous region generation method for computer-assisted pelvic tumor resection surgery was assessed by general results, oncological and functional results. Results: All patients successfully underwent surgery with a mean operative time of 252 min and average intraoperative blood loss of 1358 mL. The mean hospital stay was 22 days, and all patients completed follow-up (mean, 37 months). Two patients developed postoperative wound complications, which resolved after debridement. Adequate surgical margins were achieved in all cases. The 5-year overall survival rate was 75.6%, increasing to 80.0% among patients with wide-margin resections. At the final follow-up, the mean MSTS score among 16 limb-salvage patients was 26.6, corresponding to an average functional recovery of 88.5%. Most patients exhibited a normal gait and were able to ambulate without assistive devices. Conclusions: This dangerous region generation method, when combined with computer-assisted techniques for pelvic bone tumor resection, is feasible and can achieve favorable clinical outcomes.

Background/Objectives: Dynamic computer-assisted surgery (dCAS) has emerged as a promising tool, particularly in implantology, enabling real-time procedural adjustments through 3D image-based tracking. However, their application in other areas of oral surgery remains limited. This systematic review aims to evaluate the advantages, limitations, clinical implications, and complications associated with the use of dCAS in oral surgery (excluding implants or miniscrew insertion) beyond implant placement, in comparison to conventional freehand (FH) techniques. Methods: A systematic review was conducted in accordance with the PRISMA guidelines. A focused PICO question was developed, and a comprehensive literature search was performed in PubMed, Scopus, and the Cochrane Library between February and March 2025, and supplemented by manual screening. The risk of bias of the included studies was evaluated using the Cochrane Risk of Bias tool (RoB 2) for randomized controlled trials (RCTs) and the ROBINS-I tool for non-randomized controlled trials (NRCTs). Data were summarized in tables and analyzed through qualitative synthesis. Results: Ten studies evaluating dCAS in several oral surgical procedures, including complex tooth extractions and endodontic surgery, were included. A substantial improvement was observed in accuracy of endodontic procedures. Operator experience was a key factor in surgical outcomes. Regarding postoperative complications, no significant differences were observed, although the trend indicated an equal or lower risk in comparison with conventional FH techniques. Conclusions: dCAS may significantly improve accuracy and efficiency in endodontic surgery and reduce operative time in complex mandibular third molar (M3M) extractions. The complication rate is comparable to that of conventional FH techniques. However, current evidence remains limited, heterogeneous, and mainly experimental. Further studies are recommended to validate the benefits of dCAS in clinical settings.

Background: Reconstruction of the maxilla and midface remains one of the most demanding challenges in craniofacial surgery, requiring precise planning and a clear understanding of defect geometry to achieve functional and esthetic restoration. Advances in computer-assisted surgery (CAS) and virtual surgical planning (VSP), based on 3D segmentation of radiologic imaging, have significantly improved the management of maxillary deformities, allowing for further knowledge of patient-specific information, including anatomy, pathology, surgical planning, and reconstructive issues. The integration of computer-aided design and manufacturing (CAD/CAM) and 3D printing has further transformed reconstruction through customized titanium meshes, implants, and surgical guides. Methods:This systematic review, conducted following PRISMA 2020 guidelines, synthesizes evidence from clinical studies on CAD/CAM-assisted reconstruction of maxillary and midfacial defects of congenital, acquired, or post-resection origin. It highlights the advantages and drawbacks of maxillary reconstruction with patient-specific implants (PSISs). Primary outcomes are represented by accuracy in VSP reproduction, while secondary outcomes included esthetic results, functions, and assessment of complications. Results: Of the 44 identified articles, 10 met inclusion criteria with a time frame from April 2013 to July 2022. The outcomes of 24 treated patients are reported. CAD/CAM-guided techniques seemed to improve osteotomy accuracy, flap contouring, and implant adaptation. Conclusions: Although current data support the efficacy and safety of CAD/CAM-based approaches, limitations persist, including high costs, technological dependency, and variable long-term outcome data. This article critically evaluates the role of PSISs in maxillofacial reconstruction and outlines future directions for its standardization and broader adoption in clinical practice.

Anatomic total shoulder arthroplasty (TSA) has advanced considerably in recent decades, with the glenoid component recognized as a pivotal determinant of patient outcomes. This review integrates current literature and biomechanical insights surrounding contemporary glenoid implant designs, including all-polyethylene, hybrid, metal-backed, augmented, and inlay options. It examines key considerations such as the balance between conforming and nonconforming articulations, differences between keeled and pegged fixation, and how implant characteristics affect micromotion, radiolucency, and loosening risks. Additionally, innovative technologies like patient-specific instrumentation, computer-assisted surgery, and artificial intelligence are discussed in relation to improving implant positioning and surgical precision. By collating available evidence and implant strategies for glenoid replacement in anatomic TSA, this article aims to assist orthopaedic surgeons in making informed, patient-specific implant choices based on individual anatomy and functional requirements.

Impact of the surgical workflow and technology change on early clinical outcomes in total knee arthroplasty.

Attention-aware network with lightness embedding and Hybrid Guided Embedding for laparoscopic image desmoking.

Computer-assisted surgery (CAS) has transformed jaw reconstruction, enabling one-stage functional reconstruction with fibula bone segmentation and dental implantation. However, virtual surgical planning (VSP) without consideration of perforator position and soft tissue can lead to a difficult flap inset. Despite the rapid development of bony planning in recent years, little progress has been made in planning soft tissue, which limits the predictable execution of the plan.An artificial intelligence-enabled automatic program for segmenting skin perforating vessels was developed. The models of skin perforating vessels were incorporated into the VSP workflow when designing osteotomy and dental implantation sites. During the surgery, the vessel models were automatically registered and projected onto the patient's skin for goggle-free augmented-reality (AR) flap design and harvest. Consecutive patients undergoing FFF reconstructions participated in this pilot study. The perforators identified by the AI program and during surgery were compared.Seventy-nine skin perforators in 26 consecutive patients were mapped out using the AI program and incorporated into the routine VSP workflow. No intraoperative adjustments to the plan were required due to the perforator location. The overall predictive accuracy of the skin perforator segmentation was 92.8%.For the first time, our study employed a combined AI and AR approach for preoperative planning and intraoperative execution of the soft tissue component in computer-assisted jaw reconstruction, demonstrating promising outcomes. This study demonstrated the potential of AI and AR to improve the predictability of functional jaw reconstruction.

Estimating the 6 degrees of freedom (DoF) pose of an endoscope is crucial for various applications in minimally invasive computer-assisted surgery. Image-based approaches are some of the most practical solutions for pose estimation in surgical environments, due to a limited workspace and sensor constraints. However, these methods often struggle or fail in dynamic scenes, such as those involving tissue deformation, surgical tool movement, and tool-tissue interaction. We propose DyEndoVO, an end-to-end visual odometry method in dynamic endoscopic scenes. Our method consists of a transformer-based motion detection network and a weighted pose-optimization module. The motion detection network infers scene dynamics and guides the pose estimation. Furthermore, we introduce a semi-synthetic dataset featuring tissue and tool movement categories. It serves as training data, improving pose estimation accuracy, and also includes motion masks to enable a fine-grained inspection and evaluation. DyEndoVO significantly outperforms state-of-the-art methods in pose estimation for dynamic surgical scenes. Despite being trained solely on a synthetic dataset, our method generalizes well to real-world data without fine-tuning. Further analysis attributes this success to the effective detection of scene dynamics and the adaptation in the learned weight toward pose estimation; moreover, the semi-synthetic dataset also plays a key role in bridging the sim-to-real gap. In this work, we aim to improve the accuracy and robustness of pose estimation in challenging dynamic surgical scenes, by effectively handling scene dynamics. Our method, combined with the proposed synthetic dataset, demonstrates improved performance in pose estimation and generalizes well to real-world data, showing its potential in advancing related works such as SLAM and 3D reconstruction in complex surgical environments.

Fractures of the lower jaw are the most common type of damage to the bones of the facial skeleton. However, despite the introduction of new surgical technologies into clinical practice, the incidence of complications in patients with this type of injury remains unacceptably high, which dictates the need to improve their treatment methods. Objective of the study — to assess the capabilities and effectiveness of digital technologies in providing medical care to patients with fractures of the lower jaw, based on an analysis of literary data. This study is the first systematic review on this issue. The RINTS, Medline (PubMed), Google Scholar databases were studied from 2000 to 2024. Search terms were used reflecting the concepts of “digital technologies”, “fractures of the lower jaw”, “osteosynthesis”, “computer-assisted surgery”, “convolutional neural network”. The existing studies are devoted to improving diagnostic methods, creating repositories of large clinical and radiological databases with the possibility of their automated analysis, deep learning of convolutional neural networks for interpreting the obtained radiological images, technical support of surgical manipulations in order to improve the accuracy of repositioning bone fragments and their fixation. Digital technologies currently allow evidence-based assessment of the significance of certain clinical parameters when choosing treatment tactics, play an auxiliary role in assessing radiological images, increase the accuracy of the location of fixing structures and matching bone fragments, and reduce the time of surgery. The article discusses the factors that currently hinder the widespread introduction of digital technologies in the practice of treating patients with this type of injury and ways to overcome them.

Pediatric orbital tumors are rare and complex, requiring multidisciplinary care at specialized centers. Contemporary treatment paradigms emphasize centralized care delivery through experienced multidisciplinary teams to optimize patient outcomes. Recent advances in surgical planning technologies and intraoperative navigation systems have substantially enhanced surgical safety through improvement in tumor resection and reconstruction and reduction in complications, including recurrence of the lesion. Computed-aided surgical technologies enable precise virtual planning, minimally invasive approaches and more precise reconstruction methods when necessary by mean of patient-specific cutting guides, premolded orbital plates or individual patient solutions (IPS) prosthesis. Three-dimensional biomodelling visualizes tumor architecture and aids localization while preserving neurovascular structures, and real-time neuronavigation improves safety and efficacy. We conducted a retrospective analysis of 98 pediatric patients with orbital tumors treated between 2014 and 2025 at a tertiary center to evaluate the use of computed-assisted surgical technologies and the indications for treatment. Inclusion criteria comprised all cases where computer-assisted techniques were employed. Patients were classified into two groups: Group 1-intraconal or extensive periorbital lesions with eye-sparing intent treated via craniofacial approaches; Group 2-periorbital tumors with orbital wall involvement, to analyze the use of the different technologies. Data collected included tumor age, type, location, technology used, adjunctive treatments, and postoperative outcomes. Twelve patients underwent computer-assisted surgery. Technologies employed over the last six years included intraoperative navigation, 3D planning with/without tumor segmentation, orbital-wall reconstruction by mirroring, IPS or titanium mesh bending, and preoperative biomodelling. Patients were grouped by tumor location and treatment goals: Group 1-intraorbital lesions (primarily intraconal or 270-360° involvement), including one case of orbital encephalocele treated transcranially; Group 2-periorbital tumors with orbital-wall destruction, treated mainly via midfacial approaches. Intraoperative navigation was used in 10/12 cases (8/11 with tumor segmentation); in 3 cases with ill-defined margins, navigation localized residual tumor. Virtual surgery predominated in Group 2 (4 patients) and one in Group 1, combined with cutting guides for margins and Individual Prosthetic Solutions (IPS) prosthesis fitting (two patients: titanium and PEEK). In two cases, virtual plans were performed, STL models printed, and premolded titanium meshes used. No complications related to tumor persistence or orbital disturbance were observed. Advanced surgical technologies substantially enhance safety, efficiency, and outcomes in pediatric orbital tumors. Technology-assisted approaches represent a paradigm shift in this complex field. Additional studies are needed to establish evidence-based protocols for systematic integration of technology in pediatric orbital tumor management.

UltraBoneUDF: Self-supervised bone surface reconstruction from ultrasound based on neural unsigned distance functions.

To evaluate the accuracy of dental implant placement using fully guided static computer-assisted implant surgery (s-CAIS) and autonomous robotic computer-assisted implant surgery (r-CAIS) technology in patients with transcrestal sinus floor elevation. Patients with posterior teeth loss and transcrestal sinus floor elevation using s-CAIS or r-CAIS technology for implant surgery were included in this study. A total of 34 patients with 42 implants were included in the study (17 patients with 19 implants in the autonomous r-CAIS group, 17 patients with 23 implants in the fully guided s-CAIS group). Postoperative cone-beam computed tomography (CBCT) scans were used to determine the discrepancies between the planned and actually placed implants. The preoperative and postoperative CBCT were utilized to estimate the linear deviations and angular deviations in two-dimensional (2D) and three-dimensional (3D) space. A total of 42 implants were included, with significant differences between the autonomous r-CAIS group and fully guided s-CAIS group (p < 0.001). No adverse surgical events occurred. The 3D deviations at the implant platform were 0.484 ± 0.218 mm for the autonomous r-CAIS group and 1.179 ± 0.776 mm for the fully guided s-CAIS group, respectively. The mean linear deviations at the implant apex were 0.527 ± 0.247 and 1.196 ± 0.830 mm, respectively. The mean angular deviation was 0.882° ± 0.967° for the autonomous r-CAIS group and 2.478° ± 1.524° for the fully guided s-CAIS group. Autonomous r-CAIS technology provided a more accurate surgical approach for implant placement in patients with transcrestal sinus floor elevation than fully guided s-CAIS.

Mandibular reconstruction following segmental resection, often necessitated by oral malignancies, is a complex procedure that markedly affects postoperative function and aesthetics. Computer-assisted surgery (CAS) has emerged as a valuable tool for enhancing surgical precision in oncological head and neck reconstruction. The present study evaluated the accuracy of mandibular reconstruction using three approaches: Conventional (non-CAS), outsourced CAS (TruMatch) and an in-house CAS system developed at The University of Osaka Graduate School of Dentistry (Suita, Japan). A total of 32 cases, predominantly involving malignant tumors, were retrospectively analyzed using three-dimensional model overlay techniques to compare pre- and postoperative outcomes. While both CAS methods significantly improved accuracy compared with the non-CAS approach, no statistical difference was found between the outsourced and in-house CAS groups. These findings suggested that in-house CAS systems can achieve comparable precision to commercial solutions, offering a cost-effective and accessible alternative for advanced mandibular reconstruction in oncology settings. The results support broader adoption of in-house CAS approaches, particularly in time-sensitive or resource-limited cancer surgeries.

IntroductionIn computer-assisted bone tumor resection, surgeons manually plan cut planes with a safe margin before surgery and follow them using navigation during osteotomy. However, manual planning is prone to error and often fails to ensure adequate margins. To address this, we propose an efficient method to rapidly and accurately generate a 3D safe-margin volume that uniformly extends the tumor by a safe margin.MethodsThe study was conducted on 20 patients (9 males, 11 females) between May 2018 and October 2023. The average age was 41.75 ± 14.72 years (14–66) and the tumor types were chondrosarcoma in 5 cases, giant cell tumor in 5 cases, osteosarcoma in 2 cases, chordoma in 2 cases, Ewing sarcoma in 2 cases, spindle cell sarcoma in 1 case, osteochondroma in 1 case, chondromyxoid fibroma in 1 case and peripheral nerve sheath tumor in 1 case. The quality of the generated safe-margin volumes were assessed by visual comparison outcomes, geometric errors and maximum absolute geometric errors, time costs and clinical outcomes.ResultsAll 20 patients were successfully followed up, with a mean follow-up duration of 42.30 ± 18.75 months (range: 3–86 months). The generated 3D safe-margin volumes were visually closer to the ground truth compared to those from the 3D morphological dilation and anisotropic distance transform methods. The method achieved a mean geometric error of approximately 0.10 mm, significantly lower than the dilation method (up to 10.00 mm) and the anisotropic method (about 1.00 mm). The average maximum absolute geometric error was 0.1818 mm, and statistical tests confirmed significant improvements (P-value < 0.01). The method also exhibited favorable computational efficiency, with an average runtime of 26.82 s, substantially faster than our previous point-based method (6.29 min) and acceptable for preoperative planning.ConclusionIn this study, we developed a fast and accurate safe-margin volume generation method by combining 3D image resampling with anisotropic distance transform. This method shows strong potential in clinical practice.

Margin control is a crucial prognostic factor in head and neck oncological surgery. This retrospective case-control study aims to assess the superiority of computer-assisted surgery compared to traditional surgery in achieving optimal resection margins in maxillofacial oncologic surgery. Eighty patients with stage T3 or T4 oral squamous cell carcinoma were included, equally divided into computer-assisted surgery (CAS) and freehand groups. Negative, close, and positive margin rates were compared. Logistic regression with Firth's correction and multivariable Cox models for disease-free survival (DFS) and overall survival (OS) were applied, with tobacco and alcohol included as covariates. Sensitivity analyses used inverse probability of treatment weighting (IPTW)-weighted models. CAS was associated with a significantly lower risk of positive margins (OR 0.24; 95% CI: 0.06-0.78; p = 0.01), confirmed in the IPTW analysis. DFS was significantly improved with CAS in both multivariable and IPTW Cox models (HR 0.41 and 0.46, respectively), while OS showed no significant benefit. Nodal status emerged as the strongest prognostic factor. Despite its univocality, the CAS technology seems to offer significant advantages in achieving precise oncological margin control for oral squamous cell carcinomas.

ObjectiveRegistering a preoperative 3D model of an organ with its actual anatomy viewed from an intraoperative video is a fundamental challenge in computer-assisted surgery, especially for surgical augmented reality. To address this, we present a benchmark of state-of-the-art deep learning point-cloud registration methods, offering a transparent evaluation of their generalizability to surgical scenarios and establishing a robust guideline for developing advanced non-rigid algorithms.MethodsWe systematically evaluate traditional and deep learning GMM-based, correspondence-based, correspondence-free, matching-based, and liver-specific point cloud registration approaches on two surgical datasets: a deformed IRCAD liver set and DePoll dataset. We also propose our complete-to-partial point cloud registration framework that leverages keypoint extraction, overlap estimation, and a Transformer-based architecture, culminating in competitive registration results.ResultsExperimental evaluations on deformed IRCAD tests reveal that most deep learning methods achieve good registration performances with TRE<10 mm, MAE(R) < 4 and MAE(t)<5 mm. On DePoll, however, performance drops dramatically due to the large deformations.ConclusionIn conclusion, deep-learning rigid registration methods remain reliable under small deformations and varying partiality but lose accuracy when faced with severe non-rigid changes. To overcome this, future work should focus on building non-rigid registration architectures that preserve the strengths of self-, cross-attention and overlap modules while enhancing correspondence estimation to handle large deformations in laparoscopic surgery.

A single hip-knee-ankle angle (HKA) angle does not reflect the biomechanics of native, arthritic or prosthetic knees. Since HKA varies throughout flexion, dynamic coronal alignment is best represented by a kinematic curve plotting HKA against the range of motion. This study aimed to evaluate the relationship between kinematic HKA curves and the bony morphology of the distal femur and proximal tibia. We hypothesised that variations in distal femoral and proximal tibial anatomy are associated with distinct patterns of dynamic coronal alignment. This was an experimental study using a non-weight-bearing articulated surgical education bone model including hemipelvis, femur and tibia. Articular surfaces and bony landmarks were registered with a computer navigation system. Using medial opening wedge femoral and tibial osteotomies and a rotational femoral osteotomy, 70 morphotypes were created by altering distal femoral angle (DFA), proximal tibial angle (PTA) and femoral axial angle (FAA). For each configuration, a coronal kinematic curve was recorded from 0° to 120° of flexion. Five curve morphotypes were identified: straight, drift, inverse drift, C-shaped and inverse C-shaped. DFA and FAA differed significantly among morphotypes (p < 0.001), whereas PTA had no effect (p = 0.084). Paired comparisons confirmed significant differences in DFA and FAA across curve types. In this sawbone model, dynamic coronal plane alignment curve morphology was determined by distal femoral coronal and torsional anatomy, while tibial anatomy shifted the curve position without altering morphology. Restoring the pre-arthritic curve in TKA requires restoring DFA and FAA, whereas achieving a neutral straight curve requires individualised FAA adjustment. Consistently producing a neutral straight curve is not possible without computer-assisted or robotic surgery. These findings require validation in cadaveric or clinical studies but may guide surgical strategies aiming to reproduce native knee kinematics. N/A.

In augmented reality (AR)-guided surgical navigation, preoperative organ models are superimposed onto the patient's intraoperative anatomy to visualize critical structures such as vessels and tumors. Accurate deformation modeling is essential to maintain the reliability of AR overlays by ensuring alignment between preoperative models and the dynamically changing anatomy. Although the finite element method (FEM) offers physically plausible modeling, its high computational cost limits intraoperative applicability. Moreover, existing algorithms often fail to handle large anatomical changes, such as those induced by pneumoperitoneum or ligament dissection, leading to inaccurate anatomical correspondences and compromised AR guidance. To address these challenges, we propose a data-driven biomechanics algorithm that preserves FEM-level accuracy while improving computational efficiency. In addition, we introduce a novel human-in-the-loop mechanism into the deformation modeling process. This enables surgeons to interactively provide prompts to correct anatomical misalignments, thereby incorporating clinical expertise and allowing the model to adapt dynamically to complex surgical scenarios. Experiments on a publicly available dataset demonstrate that our algorithm achieves a mean target registration error of 3.42 mm. Incorporating surgeon prompts through the interactive framework further reduces the error to 2.78 mm, surpassing state-of-the-art methods in volumetric accuracy. These results highlight the ability of our framework to deliver efficient and accurate deformation modeling while enhancing surgeon-algorithm collaboration, paving the way for safer and more reliable computer-assisted surgeries.