Correspondence attention for facial appearance simulation.

In orthognathic surgical planning for patients with jaw deformities, it is crucial to accurately simulate the changes in facial appearance that follow the bony movement. Compared with the traditional biomechanics-based methods like the finite-element method (FEM), which are both labor-intensive and computationally inefficient, deep learning-based methods offer an efficient and robust modeling alternative. However, current methods do not account for the physical relationship between facial soft tissue and bony structure, causing them to fall short in accuracy compared to FEM. In this work, we propose an Attentive Correspondence assisted Movement Transformation network (ACMT-Net) to predict facial changes by correlating facial soft tissue changes with bony movement through a point-to-point attentive correspondence matrix. To ensure efficient training, we also introduce a contrastive loss for self-supervised pre-training of the ACMT-Net with a k-Nearest Neighbors (k-NN) based clustering. Experimental results on patients with jaw deformities show that our proposed solution can achieve significantly improved computational efficiency over the state-of-the-art FEM-based method with comparable facial change prediction accuracy.

Simulation of Postoperative Facial Appearances via Geometric Deep Learning for Efficient Orthognathic Surgical Planning.

Orthognathic surgery corrects jaw deformities to improve aesthetics and functions. Due to the complexity of the craniomaxillofacial (CMF) anatomy, orthognathic surgery requires precise surgical planning, which involves predicting postoperative changes in facial appearance. To this end, most conventional methods involve simulation with biomechanical modeling methods, which are labor intensive and computationally expensive. Here we introduce a learning-based framework to speed up the simulation of postoperative facial appearances. Specifically, we introduce a facial shape change prediction network (FSC-Net) to learn the nonlinear mapping from bony shape changes to facial shape changes. FSC-Net is a point transform network weakly-supervised by paired preoperative and postoperative data without point-wise correspondence. In FSC-Net, a distance-guided shape loss places more emphasis on the jaw region. A local point constraint loss restricts point displacements to preserve the topology and smoothness of the surface mesh after point transformation. Evaluation results indicate that FSC-Net achieves 15× speedup with accuracy comparable to a state-of-the-art (SOTA) finite-element modeling (FEM) method.

In vivo measurement of intradiscal pressure changes related to thrust and non-thrust spinal manipulation in an animal model: a pilot study.

BACKGROUND: The intervertebral disc is a known back pain generator and is frequently the focus of spinal manipulative therapy evaluation and treatment. The majority of our current knowledge regarding intradiscal pressure (IDP) changes related to spinal manual therapy involves cadaveric studies with their inherent limitations. Additional in vivo animal models are needed to investigate intervertebral disc physiological and molecular mechanisms related to spinal manipulation and spinal mobilization treatment for low back disorders. METHODS: Miniature pressure catheters (Millar SPR-1000) were inserted into either the L4-L5 or L5-L6 intervertebral disc of 3 deeply anesthetized adult cats (Oct 2012-May 2013). Changes in IDP were recorded during delivery of instrument-assisted spinal manipulation (Activator V® and Pulstar®) and motorized spinal flexion with/without manual spinous process contact. RESULTS: Motorized flexion of 30° without spinous contact decreased IDP of the L4-L5 disc by ~ 2.9 kPa, while physical contact of the L4 spinous process decreased IDP an additional ~ 1.4 kPa. Motorized flexion of 25° with L5 physical contact in a separate animal decreased IDP of the L5-L6 disc by ~ 1.0 kPa. Pulstar® impulses (setting 1-3) increased IDP of L4-L5 and L5-L6 intervertebral discs by ~ 2.5 to 3.0 kPa. Activator V® (setting 1-4) impulses increased L4-L5 IDP to a similar degree. Net changes in IDP amplitudes remained fairly consistent across settings on both devices regardless of device setting suggesting that viscoelastic properties of in vivo spinal tissues greatly dampen superficially applied manipulative forces prior to reaching deep back structures such as the intervertebral disc. CONCLUSIONS: This study marks the first time that feline in vivo changes in IDP have been reported using clinically available instrument-assisted spinal manipulation devices and/or spinal mobilization procedures. The results of this pilot study indicate that a feline model can be used to investigate IDP changes related to spinal manual therapy mechanisms as well as the diminution of these spinal manipulative forces due to viscoelastic properties of the surrounding spinal tissues. Additional investigation of IDP changes is warranted in this and/or other in vivo animal models to provide better insights into the physiological effects and mechanisms of spinal manual therapy at the intervertebral disc level.

Measurement of Force and Intramuscular Pressure Changes Related to Thrust Spinal Manipulation in an In Vivo Animal Model.

Current knowledge regarding biomechanical in vivo deep tissue measures related to spinal manipulation remain somewhat limited. More in vivo animal studies are needed to better understand the effects viscoelastic tissue properties (i.e., dampening) have on applied spinal manipulation forces. This new knowledge may eventually help to determine whether positive clinical outcomes are associated with particular force thresholds reaching superficial and/or deep spinal tissues. A computer-controlled feedback motor and a modified Activator V device with a dynamic load cell attached were used to deliver thrust spinal manipulations at various magnitudes to the L7 spinous process in deeply anesthetized animals. Miniature pressure catheters (Millar SPR-1000) were inserted unilaterally into superficial and deep multifidi muscles. Measurements of applied mechanical forces and superficial/deep multifidi intramuscular pressure changes were recorded during spinal manipulations delivered in vivo. Manipulative forces and net changes in intramuscular pressures reaching deep spinal tissues are greatly diminished by viscoelastic properties of in vivo tissues, which could have possible clinical safety and/or mechanistic implications.

Deep Simulation of Facial Appearance Changes Following Craniomaxillofacial Bony Movements in Orthognathic Surgical Planning.

Facial appearance changes with the movements of bony segments in orthognathic surgery of patients with craniomaxillofacial (CMF) deformities. Conventional bio-mechanical methods, such as finite element modeling (FEM), for simulating such changes, are labor intensive and computationally expensive, preventing them from being used in clinical settings. To overcome these limitations, we propose a deep learning framework to predict post-operative facial changes. Specifically, FC-Net, a facial appearance change simulation network, is developed to predict the point displacement vectors associated with a facial point cloud. FC-Net learns the point displacements of a pre-operative facial point cloud from the bony movement vectors between pre-operative and simulated post-operative bony models. FC-Net is a weakly-supervised point displacement network trained using paired data with strict point-to-point correspondence. To preserve the topology of the facial model during point transform, we employ a local-point-transform loss to constrain the local movements of points. Experimental results on real patient data reveal that the proposed framework can predict post-operative facial appearance changes remarkably faster than a state-of-the-art FEM method with comparable prediction accuracy.

A novel incremental simulation of facial changes following orthognathic surgery using FEM with realistic lip sliding effect.

Accurate prediction of facial soft-tissue changes following orthognathic surgery is crucial for surgical outcome improvement. We developed a novel incremental simulation approach using finite element method (FEM) with a realistic lip sliding effect to improve the prediction accuracy in the lip region. First, a lip-detailed mesh is generated based on accurately digitized lip surface points. Second, an improved facial soft-tissue change simulation method is developed by applying a lip sliding effect along with the mucosa sliding effect. Finally, the orthognathic surgery initiated soft-tissue change is simulated incrementally to facilitate a natural transition of the facial change and improve the effectiveness of the sliding effects. Our method was quantitatively validated using 35 retrospective clinical data sets by comparing it to the traditional FEM simulation method and the FEM simulation method with mucosa sliding effect only. The surface deviation error of our method showed significant improvement in the upper and lower lips over the other two prior methods. In addition, the evaluation results using our lip-shape analysis, which reflects clinician's qualitative evaluation, also proved significant improvement of the lip prediction accuracy of our method for the lower lip and both upper and lower lips as a whole compared to the other two methods. In conclusion, the prediction accuracy in the clinically critical region, i.e., the lips, significantly improved after applying incremental simulation with realistic lip sliding effect compared with the FEM simulation methods without the lip sliding effect.

A New Approach of Predicting Facial Changes following Orthognathic Surgery using Realistic Lip Sliding Effect.

Accurate prediction of facial soft-tissue changes following orthognathic surgery is crucial for improving surgical outcome. However, the accuracy of current prediction methods still requires further improvement in clinically critical regions, especially the lips. We develop a novel incremental simulation approach using finite element method (FEM) with realistic lip sliding effect to improve the prediction accuracy in the area around the lips. First, lip-detailed patient-specific FE mesh is generated based on accurately digitized lip surface landmarks. Second, an improved facial soft-tissue change simulation method is developed by applying a lip sliding effect in addition to the mucosa sliding effect. The soft-tissue change is then simulated incrementally to facilitate a natural transition of the facial change and improve the effectiveness of the sliding effects. A preliminary evaluation of prediction accuracy was conducted using retrospective clinical data. The results showed that there was a significant prediction accuracy improvement in the lip region when the realistic lip sliding effect was applied along with the mucosa sliding effect.

An eFTD-VP framework for efficiently generating patient-specific anatomically detailed facial soft tissue FE mesh for craniomaxillofacial surgery simulation.

Accurate surgical planning and prediction of craniomaxillofacial surgery outcome requires simulation of soft tissue changes following osteotomy. This can only be achieved by using an anatomically detailed facial soft tissue model. The current state-of-the-art of model generation is not appropriate to clinical applications due to the time-intensive nature of manual segmentation and volumetric mesh generation. The conventional patient-specific finite element (FE) mesh generation methods are to deform a template FE mesh to match the shape of a patient based on registration. However, these methods commonly produce element distortion. Additionally, the mesh density for patients depends on that of the template model. It could not be adjusted to conduct mesh density sensitivity analysis. In this study, we propose a new framework of patient-specific facial soft tissue FE mesh generation. The goal of the developed method is to efficiently generate a high-quality patient-specific hexahedral FE mesh with adjustable mesh density while preserving the accuracy in anatomical structure correspondence. Our FE mesh is generated by eFace template deformation followed by volumetric parametrization. First, the patient-specific anatomically detailed facial soft tissue model (including skin, mucosa, and muscles) is generated by deforming an eFace template model. The adaptation of the eFace template model is achieved by using a hybrid landmark-based morphing and dense surface fitting approach followed by a thin-plate spline interpolation. Then, high-quality hexahedral mesh is constructed by using volumetric parameterization. The user can control the resolution of hexahedron mesh to best reflect clinicians' need. Our approach was validated using 30 patient models and 4 visible human datasets. The generated patient-specific FE mesh showed high surface matching accuracy, element quality, and internal structure matching accuracy. They can be directly and effectively used for clinical simulation of facial soft tissue change.

A clinically validated prediction method for facial soft-tissue changes following double-jaw surgery.

PURPOSE: It is clinically important to accurately predict facial soft-tissue changes prior to orthognathic surgery. However, the current simulation methods are problematic, especially in anatomic regions of clinical significance, e.g., the nose, lips, and chin. We developed a new 3-stage finite element method (FEM) approach that incorporates realistic tissue sliding to improve such prediction. METHODS: In Stage One, soft-tissue change was simulated, using FEM with patient-specific mesh models generated from our previously developed eFace template. Postoperative bone movement was applied on the patient mesh model with standard FEM boundary conditions. In Stage Two, the simulation was improved by implementing sliding effects between gum tissue and teeth using a nodal force constraint scheme. In Stage Three, the result of the tissue sliding effect was further enhanced by reassigning the soft-tissue-bone mapping and boundary conditions using nodal spatial constraint. Finally, our methods have been quantitatively and qualitatively validated using 40 retrospectively evaluated patient cases by comparing it to the traditional FEM method and the FEM with sliding effect, using a nodal force constraint method. RESULTS: The results showed that our method was better than the other two methods. Using our method, the quantitative distance errors between predicted and actual patient surfaces for the entire face and any subregions thereof were below 1.5 mm. The overall soft-tissue change prediction was accurate to within 1.1 ± 0.3 mm, with the accuracy around the upper and lower lip regions of 1.2 ± 0.7 mm and 1.5 ± 0.7 mm, respectively. The results of qualitative evaluation completed by clinical experts showed an improvement of 46% in acceptance rate compared to the traditional FEM simulation. More than 80% of the result of our approach was considered acceptable in comparison with 55% and 50% following the other two methods. CONCLUSION: The FEM simulation method with improved sliding effect showed significant accuracy improvement in the whole face and the clinically significant regions (i.e., nose and lips) in comparison with the other published FEM methods, with or without sliding effect using a nodal force constraint. The qualitative validation also proved the clinical feasibility of the developed approach.

Characteristics of Paraspinal Muscle Spindle Response to Mechanically Assisted Spinal Manipulation: A Preliminary Report.

OBJECTIVES: The purpose of this preliminary study is to determine muscle spindle response characteristics related to the use of 2 solenoid powered clinical mechanically assisted manipulation (MAM) devices. METHODS: L6 muscle spindle afferents with receptive fields in paraspinal muscles were isolated in 6 cats. Neural recordings were made during L7 MAM thrusts using the Activator V (Activator Methods Int. Ltd., Phoenix, AZ) and/or Pulstar (Sense Technology Inc., Pittsburgh, PA) devices at their 3 lowest force settings. Mechanically assisted manipulation response measures included (a) the time required post-thrust until the first action potential, (b) differences in mean frequency (MF) and mean instantaneous frequency (MIF) 2 seconds before and after MAM, and (c) the time required for muscle spindle discharge (MF and MIF) to return to 95% of baseline after MAM. RESULTS: Depending on device setting, between 44% to 80% (Pulstar) and 11% to 63% (Activator V) of spindle afferents required >6 seconds to return to within 95% of baseline MF values; whereas 66% to 89% (Pulstar) and 75% to 100% (Activator V) of spindle responses returned to within 95% of baseline MIF in <6 seconds after MAM. Nonparametric comparisons between the 22 N and 44 N settings of the Pulstar yielded significant differences for the time required to return to baseline MF and MIF. CONCLUSION: Short duration (<10 ms) MAM thrusts decrease muscle spindle discharge with a majority of afferents requiring prolonged periods (>6 seconds) to return to baseline MF activity. Physiological consequences and clinical relevance of described MAM mechanoreceptor responses will require additional investigation.

An eFace-Template Method for Efficiently Generating Patient-Specific Anatomically-Detailed Facial Soft Tissue FE Models for Craniomaxillofacial Surgery Simulation.

Accurate surgical planning and prediction of craniomaxillofacial surgery outcome requires simulation of soft-tissue changes following osteotomy. This can only be accomplished on an anatomically-detailed facial soft tissue model. However, current anatomically-detailed facial soft tissue model generation is not appropriate for clinical applications due to the time intensive nature of manual segmentation and volumetric mesh generation. This paper presents a novel semi-automatic approach, named eFace-template method, for efficiently and accurately generating a patient-specific facial soft tissue model. Our novel approach is based on the volumetric deformation of an anatomically-detailed template to be fitted to the shape of each individual patient. The adaptation of the template is achieved by using a hybrid landmark-based morphing and dense surface fitting approach followed by a thin-plate spline interpolation. This methodology was validated using 4 visible human datasets (regarded as gold standards) and 30 patient models. The results indicated that our approach can accurately preserve the internal anatomical correspondence (i.e., muscles) for finite element modeling. Additionally, our hybrid approach was able to achieve an optimal balance among the patient shape fitting accuracy, anatomical correspondence and mesh quality. Furthermore, the statistical analysis showed that our hybrid approach was superior to two previously published methods: mesh-matching and landmark-based transformation. Ultimately, our eFace-template method can be directly and effectively used clinically to simulate the facial soft tissue changes in the clinical application.

Neural Response During a Mechanically Assisted Spinal Manipulation in an Animal Model: A Pilot Study.

INTRODUCTION: Mechanoreceptor stimulation is theorized to contribute to the therapeutic efficacy of spinal manipulation. Use of mechanically-assisted spinal manipulation (MA-SM) devices is increasing among manual therapy clinicians worldwide. The purpose of this pilot study is to determine the feasibility of recording in vivo muscle spindle responses during a MA-SM in an intervertebral fixated animal model. METHODS: Intervertebral fixation was created by inserting facet screws through the left L5-6 and L6-7 facet joints of a cat spine. Three L6muscle spindle afferents with receptive fields in back muscles were isolated. Recordings were made during MA-SM thrusts delivered to the L7 spinous process using an instrumented Activator IV clinical device. RESULTS: Nine MA-SM thrusts were delivered with peak forces ranging from 68-122N and with thrust durations of less than 5ms. High frequency muscle spindle discharge occurred during MA-SM. Following the MA-SM, muscle spindle responses included returning to pre-manipulation levels, slightly decreasing for a short window of time, and greatly decreasing for more than 40s. CONCLUSION: This study demonstrates that recording in vivo muscle spindle response using clinical MA-SM devices in an animal model is feasible. Extremely short duration MA-SM thrusts (<5ms) can have an immediate and/or a prolonged (> 40s) effect on muscle spindle discharge. Greater peak forces during MA-SM thrusts may not necessarily yield greater muscle spindle responses. Determining peripheral response during and following spinal manipulation may be an important step in optimizing its' clinical efficacy. Future studies may investigate the effect of thrust dosage and magnitude.

In vitro biomechanical evaluation of single impulse and repetitive mechanical shockwave devices utilized for spinal manipulative therapy.

Mechanical shockwave therapy devices have been in clinical use for almost 40 years. While most often used to treat back pain, our understanding of their biomechanical performance is very limited. From biomechanical studies we know that biological tissue is viscoelastic and preferably excited around its resonance frequency. Targeting these frequencies has been the focus in extracorporeal shock wave lithotripsy, but these concepts are relatively new in orthopedic and rehabilitation therapies. The exact mechanism by which shockwave therapy acts is not known. Knowledge of the performance characteristics of these devices, correlated with clinical outcome studies, may lead to better patient selection, improvement of device functionality, and knowledge of the underlying working principals of therapy. The objectives of this study were to determine the ability of several commercial shockwave devices to achieve a desired thrust profile in a benchtop setting, determine the thrust profile in a clinical analog, and determine the influence of operator experience level on device performance. We conducted two different types of testing: (1) bench testing to evaluate the devices themselves, and (2) clinical equivalent testing to determine the influence of the operator. The results indicated a significant dependence of thrust output on the compliance of the test media. The Activator V-E device matched the ideal half-sine thrust profile to 94%, followed by the Impulse device (84%), the Activator IV/FS (74%), and the Activator II (48%). While most devices deviated from the ideal profile on the return path, the Impulse device exhibited a secondary peak. Moreover, the Activator V-E device provided evidence that the device performs consistently despite operator experience level. This has been a major concern in manual spinal manipulation. Based on our results, a hyper-flexible spine would receive a lower peak thrust force than a hypo-flexible spine at the same power setting. Furthermore, a hand-held operation further reduced the peak thrust force as it increased the system compliance. However, that influence was dissimilar for the different devices. Although controlled clinical trials are needed to determine the correlation between thrust profile and clinical outcome, already ongoing clinical studies indicate an improved patient satisfaction due to reduced treatment pain when devices are used with a thrust characteristic closer to an ideal sine wave.

Computer-aided tissue engineering: benefiting from the control over scaffold micro-architecture.

Minimization schema in nature affects the material arrangements of most objects, independent of scale. The field of cellular solids has focused on the generalization of these natural architectures (bone, wood, coral, cork, honeycombs) for material improvement and elucidation into natural growth mechanisms. We applied this approach for the comparison of a set of complex three-dimensional (3D) architectures containing the same material volume but dissimilar architectural arrangements. Ball and stick representations of these architectures at varied material volumes were characterized according to geometric properties, such as beam length, beam diameter, surface area, space filling efficiency, and pore volume. Modulus, deformation properties, and stress distributions as contributed solely by architectural arrangements was revealed through finite element simulations. We demonstrated that while density is the greatest factor in controlling modulus, optimal material arrangement could result in equal modulus values even with volumetric discrepancies of up to 10%. We showed that at low porosities, loss of architectural complexity allows these architectures to be modeled as closed celled solids. At these lower porosities, the smaller pores do not greatly contribute to the overall modulus of the architectures and that a stress backbone is responsible for the modulus. Our results further indicated that when considering a deposition-based growth pattern, such as occurs in nature, surface area plays a large role in the resulting strength of these architectures, specifically for systems like bone. This completed study represents the first step towards the development of mathematical algorithms to describe the mechanical properties of regular and symmetric architectures used for tissue regenerative applications. The eventual goal is to create logical set of rules that can explain the structural properties of an architecture based solely upon its geometry. The information could then be used in an automatic fashion to generate patient-specific scaffolds for the treatment of tissue defects.

Scaffold pore space modulation through intelligent design of dissolvable microparticles.

The goal of this area of research is to manipulate the pore space of scaffolds through the application of an intelligent design concept on dissolvable microparticles. To accomplish this goal, we developed an efficient and repeatable process for fabrication of microparticles from multiple materials using a combination of rapid prototyping (RP) and soft lithography. Phase changed 3D printing was used to create masters for PDMS molds. A photocrosslinkable polymer was then delivered into these molds to make geometrically complex 3D microparticles. This repeatable process has demonstrated to generate the objects with greater than 95% repeatability with complete pattern transfer. This process was illustrated for three different shapes of various complexities. The shapes were based on the extrusion of 2D shapes. This may allow simplification of the fabrication process in the future combined with a direct transfer of the findings. Altering the shapes of particles used for porous scaffold fabrication will allow for tailoring of the pore shapes, and therefore their biological function within a porous tissue engineering scaffold. Through permeation experiments, we have shown that the pore geometry may alter the permeability coefficient of scaffolds while influencing mechanical properties to a lesser extent. By selecting different porogen shapes, the nutrition transport and scaffold degradation can be significantly influenced with minimal effect on the mechanical integrity of the construct. In addition, the different shapes may allow a control of drug release by modifying their surface-to-volume ratio, which could modulate drug delivery over time. While soft lithography is currently used with photolithography, its high precision is offset by high cost of production. The employment of RP to a specific resolution offers a much less expensive alternative with increased throughput due to the speed of current RP systems.

Premature adjacent vertebral fracture after vertebroplasty: a biomechanical study.

BACKGROUND: There is an increased incidence of fractures in untreated adjacent vertebrae after vertebroplasty. OBJECTIVE: To introduce unconstrained 6 degrees of freedom biomechanical testing to investigate whether vertebroplasty lowered the fracture strength of adjacent untreated vertebrae under physiological loading conditions and to describe the observed fracture pattern. METHODS: Three-level spinal segments (T10-12 and L1-3) from 6 spines were tested under unconstrained axial compression in which shear forces and torque were minimized using a 6-degrees of freedom robotic arm. Fracture initiation loads and ultimate failure loads of lumbar segments were predicted from the corresponding thoracic segments by assuming constant fracture stress along the spinal column. The predicted values were compared with postvertebroplasty experimental values of the lumbar spine segments. Plain radiographs were taken at 600-N increments to record the developing fracture pattern. RESULTS: All 6 vertebroplasty group specimens experienced reductions in fracture strengths ranging from 27.4% to 47.6% with an average decrease of 32.6% (P < .002) and reductions in ultimate failure load ranging from 1.6% to 47.3%, with an average decrease of 34.7% (P < .003) compared with predicted values from the nonvertebroplasty group. In all vertebroplasty group specimens, the superior and inferior endplates of the untreated middle vertebral body (L2) were deflected, whereas 5 of the 6 nonvertebroplasty group specimens did not show any evidence of endplate deflection. CONCLUSION: Vertebroplasty altered the load transfer along the anterior spinal column, thereby statistically significantly increasing fracture risk and ultimate failure load of the untreated adjacent vertebrae. The radiographic findings support the endplate deflection fracture mechanism as the cause of adjacent fractures after vertebroplasty.

Geometry-based algorithm for the prediction of nonpathologic mandibular movement.

PURPOSE: The goal of this study was to create a model for human mandibular movement prediction based on the geometry of the mandible. MATERIALS AND METHODS: Ten nonpathologic individuals underwent motion tracking and sagittal radiographs. From the data, a mathematical algorithm for mandibular movement prediction was developed based on mandibular geometry. The algorithm was then used to predict the mandibular movement of a cadaver subject. The algorithm was also validated in a living subject by comparing to recorded mandibular movement. RESULTS: Both mandibular movement predictions were free of bone collisions and showed mandibular movement that mimicked the in vivo situation. Comparisons between the predicted and recorded mandibular movements for the living human subject verified that the prediction model was accurate. CONCLUSIONS: The mandibular movement can be predicted based on the mandibular opening radius. The model is validated in living human subjects and shows effectiveness in predictions for cadaver models. Mandibular movement prediction may be a useful tool for physicians as well as investigators who focus on temporomandibular joint research.

Clinical feasibility of computer-aided surgical simulation (CASS) in the treatment of complex cranio-maxillofacial deformities.

PURPOSE: The purpose of this study was to establish clinical feasibility of our 3-dimensional computer-aided surgical simulation (CASS) for complex craniomaxillofacial surgery. MATERIALS AND METHODS: Five consecutive patients with complex craniomaxillofacial deformities, including hemifacial microsomia, defects after tumor ablation, and deformity after TMJ reconstruction, were used. The patients' surgical interventions were planned by using the authors' CASS planning method. Computed tomography (CT) was initially obtained. The first step of the planning process was to create a composite skull model, which reproduces both the bony structures and the dentition with a high degree of accuracy. The second step was to quantify the deformity. The third step was to simulate the entire surgery in the computer. The maxillary osteotomy was usually completed first, followed by mandibular and chin surgeries. The shape and size of the bone graft, if needed, was also simulated. If the simulated outcomes were not satisfactory, the surgical plan could be modified and simulation could be started over. The final step was to create surgical splints. Using the authors' computer-aided designing/manufacturing techniques, the surgical splints and templates were designed in the computer and fabricated by a stereolithographic apparatus. To minimize the potential risks to the patients, the surgeries were also planned following the current planning methods, and acrylic surgical splints were created as a backup plan. RESULTS: All 5 patients were successfully planned using our CASS planning method. The computer-generated surgical splints were successfully used on all patients at the time of the surgery. The backup acrylic surgical splints and plans were never used. Six-week postoperative CT scans showed the surgical plans were precisely reproduced in the operating room and the deformities were corrected as planned. CONCLUSION: The results of this study have shown the clinical feasibility of our CASS planning method. Using our CASS method, we were able to treat patients with significant asymmetries in a single operation that in the past was usually completed in 2 stages. We were also able to simulate different surgical procedures to create the appropriate plan. The computerized surgical plan was then transferred to the patient in the operating room using computer-generated surgical splints.