V2-Net: An Attention-guided Volumetric Regression Network for Tooth Landmark Localization on CT Images with Metal Artifacts.

For virtual surgical planning in orthognathic surgery, marking tooth landmarks on CT images is an important procedure. However, the manual localization procedure of tooth landmarks is time-consuming, labor-intensive, and requires expert knowledge. Also, direct and automatic tooth landmark localization on CT images is difficult because of the lower resolution and metal artifacts of dental images. The purpose of this study was to propose an attention-guided volumetric regression network (V2-Net) for accurate tooth landmark localization on CT images with metal artifacts and lower resolution. V2-Net has an attention-guided network architecture using a coarse-to-fine-attention mechanism that guided the 3D probability distribution of tooth landmark locations within anatomical structures from the coarse V-Net to the fine V-Net for more focus on tooth landmarks. In addition, we combined attention-guided learning and a 3D attention module with optimal Pseudo Huber loss to improve the localization accuracy. Our results show that the proposed method achieves state-of-the-art accuracy of 0.85 ± 0.40 mm in terms of mean radial error, outperforming previous studies. In ablation studies, we observed that the proposed attention-guided learning and a 3D attention module improved the accuracy of tooth landmark localization in CT images of lower resolution and metal artifacts. Furthermore, our method achieved 97.92% in terms of the success detection rate within the clinically accepted accuracy range of 2.0 mm.

Automatic classification of 3D positional relationship between mandibular third molar and inferior alveolar canal using a distance-aware network.

The purpose of this study was to automatically classify the three-dimensional (3D) positional relationship between an impacted mandibular third molar (M3) and the inferior alveolar canal (MC) using a distance-aware network in cone-beam CT (CBCT) images. We developed a network consisting of cascaded stages of segmentation and classification for the buccal-lingual relationship between the M3 and the MC. The M3 and the MC were simultaneously segmented using Dense121 U-Net in the segmentation stage, and their buccal-lingual relationship was automatically classified using a 3D distance-aware network with the multichannel inputs of the original CBCT image and the signed distance map (SDM) generated from the segmentation in the classification stage. The Dense121 U-Net achieved the highest average precision of 0.87, 0.96, and 0.94 in the segmentation of the M3, the MC, and both together, respectively. The 3D distance-aware classification network of the Dense121 U-Net with the input of both the CBCT image and the SDM showed the highest performance of accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve, each of which had a value of 1.00. The SDM generated from the segmentation mask significantly contributed to increasing the accuracy of the classification network. The proposed distance-aware network demonstrated high accuracy in the automatic classification of the 3D positional relationship between the M3 and the MC by learning anatomical and geometrical information from the CBCT images.

Analysis of Gyro Bias Depending on the Position of Inertial Measurement Unit in Rotational Inertial Navigation Systems.

In this paper, a calibration method for gyro bias that changes depending on the position of the IMU (inertial measurement unit) is proposed to improve the navigation performance of RLG-based RINS (ring-laser-gyro-based rotational inertial navigation system). RINS is a navigation device that compensates for the inertial sensor errors by utilizing the rotation of the IMU. In previous studies, the rotation scheme of the IMU is designed assuming that inertial sensor errors are not affected by position of the IMU. However, changes in temperature distribution, direction of gravity, and dithering according to the rotation of the IMU affect the inertial sensor errors, such as gyro bias. These errors could degrade the long-term navigation performance of RLG-based RINS. To deal with this problem, this paper proposed a compensation method of the gyro bias that changes depending on the position of the IMU. First, RINS is reviewed using a dual-axis 16-position rotation scheme and RLG. Next, the attitude error of RLG-based RINS is derived utilizing navigation equations. The effect of the gyro bias change caused by the change in the IMU attitude for the navigation performance of RINS is analyzed based on navigation equations and simulations. Finally, system-level indirect calibrations for the Z-axis up position and Z-axis down position are performed to calculate the gyro bias change caused by the IMU attitude. The accuracy of the proposed calibration method is verified by long-term navigation test. The test results show that the proposed calibration method improves the navigation performance of RINS compared with the conventional calibration method.

Canal-Net for automatic and robust 3D segmentation of mandibular canals in CBCT images using a continuity-aware contextual network.

The purpose of this study was to propose a continuity-aware contextual network (Canal-Net) for the automatic and robust 3D segmentation of the mandibular canal (MC) with high consistent accuracy throughout the entire MC volume in cone-beam CT (CBCT) images. The Canal-Net was designed based on a 3D U-Net with bidirectional convolutional long short-term memory (ConvLSTM) under a multi-task learning framework. Specifically, the Canal-Net learned the 3D anatomical context information of the MC by incorporating spatio-temporal features from ConvLSTM, and also the structural continuity of the overall MC volume under a multi-task learning framework using multi-planar projection losses complementally. The Canal-Net showed higher segmentation accuracies in 2D and 3D performance metrics (p < 0.05), and especially, a significant improvement in Dice similarity coefficient scores and mean curve distance (p < 0.05) throughout the entire MC volume compared to other popular deep learning networks. As a result, the Canal-Net achieved high consistent accuracy in 3D segmentations of the entire MC in spite of the areas of low visibility by the unclear and ambiguous cortical bone layer. Therefore, the Canal-Net demonstrated the automatic and robust 3D segmentation of the entire MC volume by improving structural continuity and boundary details of the MC in CBCT images.

A Complete Digital Workflow for Planning, Simulation, and Evaluation in Orthognathic Surgery.

The purpose of this study was to develop a complete digital workflow for planning, simulation, and evaluation for orthognathic surgery based on 3D digital natural head position reproduction, a cloud-based collaboration platform, and 3D landmark-based evaluation. We included 24 patients who underwent bimaxillary orthognathic surgery. Surgeons and engineers could share the massive image data immediately and conveniently and collaborate closely in surgical planning and simulation using a cloud-based platform. The digital surgical splint could be optimized for a specific patient before or after the physical fabrication of 3D printing splints through close collaboration. The surgical accuracy was evaluated comprehensively via the translational (linear) and rotational (angular) discrepancies between identical 3D landmarks on the simulation and postoperative computed tomography (CT) models. The means of the absolute linear discrepancy at eight tooth landmarks were 0.61 ± 0.55, 0.86 ± 0.68, and 1.00 ± 0.79 mm in left-right, advance-setback, and impaction-elongation directions, respectively, and 1.67 mm in the root mean square direction. The linear discrepancy in the left-right direction was significantly different from the other two directions as shown by analysis of variance (ANOVA, p < 0.05). The means of the absolute angular discrepancies were 1.43 ± 1.06°, 0.50 ± 0.31°, and 0.58 ± 0.41° in the pitch, roll, and yaw orientations, respectively. The angular discrepancy in the pitch orientation was significantly different from the other two orientations (ANOVA, p < 0.05). The complete digital workflow that we developed for orthognathic patients provides efficient and streamlined procedures for orthognathic surgery and shows high surgical accuracy with efficient image data sharing and close collaboration.

QCBCT-NET for direct measurement of bone mineral density from quantitative cone-beam CT: a human skull phantom study.

The purpose of this study was to directly and quantitatively measure BMD from Cone-beam CT (CBCT) images by enhancing the linearity and uniformity of the bone intensities based on a hybrid deep-learning model (QCBCT-NET) of combining the generative adversarial network (Cycle-GAN) and U-Net, and to compare the bone images enhanced by the QCBCT-NET with those by Cycle-GAN and U-Net. We used two phantoms of human skulls encased in acrylic, one for the training and validation datasets, and the other for the test dataset. We proposed the QCBCT-NET consisting of Cycle-GAN with residual blocks and a multi-channel U-Net using paired training data of quantitative CT (QCT) and CBCT images. The BMD images produced by QCBCT-NET significantly outperformed the images produced by the Cycle-GAN or the U-Net in mean absolute difference (MAD), peak signal to noise ratio (PSNR), normalized cross-correlation (NCC), structural similarity (SSIM), and linearity when compared to the original QCT image. The QCBCT-NET improved the contrast of the bone images by reflecting the original BMD distribution of the QCT image locally using the Cycle-GAN, and also spatial uniformity of the bone images by globally suppressing image artifacts and noise using the two-channel U-Net. The QCBCT-NET substantially enhanced the linearity, uniformity, and contrast as well as the anatomical and quantitative accuracy of the bone images, and demonstrated more accuracy than the Cycle-GAN and the U-Net for quantitatively measuring BMD in CBCT.

Reliable Time Propagation Algorithms for PMF and RBPMF.

This paper addresses the reliable time propagation algorithms for Point Mass Filter (PMF) and Rao-Blackwellized PMF (RBPMF) for the nonlinear estimaton problem. The conventional PMF and RBPMF process the probability diffusion for the time propagation with the direct sampled-values of the process noise. However, if the grid interval is not dense enough, it fails to represent the statistical characteristics of the noise accurately so the performance might deteriorate. To overcome that problem, we propose time propagation convolution algorithms adopting Moment Matched Gaussian Kernel (MMGK) on regular grids through mass linear interpolation. To extend the dimension of the MMGK that can accurately describe the noise moments up to the kernel length, we propose the extended MMGK based on the outer tensor product. The proposed time propagation algorithms using one common kernel through the mass linear interpolation not only improve the performance of the filter but also significantly reduce the computational load. The performance improvement and the computational load reduction of the proposed algorithms are verified through numerical simulations for various nonlinear models.

Quantitative Augmented Reality-Assisted Free-Hand Orthognathic Surgery Using Electromagnetic Tracking and Skin-Attached Dynamic Reference.

The purpose of this study was to develop a quantitative AR-assisted free-hand orthognathic surgery method using electromagnetic (EM) tracking and skin-attached dynamic reference. The authors proposed a novel, simplified, and convenient workflow for augmented reality (AR)-assisted orthognathic surgery based on optical marker-less tracking, a comfortable display, and a non-invasive, skin-attached dynamic reference frame. The 2 registrations between the physical (EM tracking) and CT image spaces and between the physical and AR camera spaces, essential processes in AR-assisted surgery, were pre-operatively performed using the registration body complex and 3D depth camera. The intraoperative model of the maxillary bone segment (MBS) was superimposed on the real patient image with the simulated goal model on a flat-panel display, and the MBS was freely handled for repositioning with respect to the skin-attached dynamic reference tool (SRT) with quantitative visualization of landmarks of interest using only EM tracking. To evaluate the accuracy of AR-assisted Le Fort I surgery, the MBS of the phantom was simulated and repositioned by 6 translational and three rotational movements. The mean absolute deviations (MADs) between the simulation and post-operative positions of MBS landmarks by the SRT were 0.20, 0.34, 0.29, and 0.55 mm in x- (left lateral, right lateral), y- (setback, advance), and z- (impaction, elongation) directions, and RMS, respectively, while those by the BRT were 0.23, 0.37, 0.30, and 0.60 mm. There were no significant differences between the translation and rotation surgeries or among surgeries in the x-, y-, and z-axes for the SRT. The MADs in the x-, y-, and z-axes exhibited no significant differences between the SRT and BRT. The developed method showed high accuracy and reliability in free-hand orthognathic surgery using EM tracking and skin-attached dynamic reference.

Development of panorama resolution phantom for comprehensive evaluation of the horizontal and vertical resolution of panoramic radiography.

Panoramic radiography is the most commonly used equipment in the dental field, but there is no comprehensive standard about how to evaluate the spatial resolution of panoramic radiography. In this study, panorama resolution phantoms were developed for evaluation of horizontal and vertical resolution reflecting unique characteristics of panoramic radiography. Four horizontal resolution phantoms of staircase shape were designed to obtain images of horizontal lead line pairs in a 52 mm width. Four vertical resolution phantoms with vertical lead line pairs placed at an oblique angle were also designed. A phantom stand was made. Three machines were evaluated twice by two oromaxillofacial radiologists. The horizontal lead line pairs were readable over the entire measured area at the values of 1.88, 2.32, and 2.58 lp/mm for all machines. A readable area of horizontal 3.19 lp/mm was observed in the lingual side. In the vertical resolution phantoms, it was possible to read a narrower range. By using the panorama resolution phantoms and phantom stand, it was possible to evaluate the resolution of a wide buccolingual width in four significant areas. By evaluating the resolution of the vertical and horizontal compartments separately, it was possible to gain a better understanding of the obtained images.

Deep Learning Hybrid Method to Automatically Diagnose Periodontal Bone Loss and Stage Periodontitis.

We developed an automatic method for staging periodontitis on dental panoramic radiographs using the deep learning hybrid method. A novel hybrid framework was proposed to automatically detect and classify the periodontal bone loss of each individual tooth. The framework is a hybrid of deep learning architecture for detection and conventional CAD processing for classification. Deep learning was used to detect the radiographic bone level (or the CEJ level) as a simple structure for the whole jaw on panoramic radiographs. Next, the percentage rate analysis of the radiographic bone loss combined the tooth long-axis with the periodontal bone and CEJ levels. Using the percentage rate, we could automatically classify the periodontal bone loss. This classification was used for periodontitis staging according to the new criteria proposed at the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. The Pearson correlation coefficient of the automatic method with the diagnoses by radiologists was 0.73 overall for the whole jaw (p < 0.01), and the intraclass correlation value 0.91 overall for the whole jaw (p < 0.01). The novel hybrid framework that combined deep learning architecture and the conventional CAD approach demonstrated high accuracy and excellent reliability in the automatic diagnosis of periodontal bone loss and staging of periodontitis.

Method for Improved Performance of Fixed-Gain Self-Alignment in the Temperature Stabilizing State.

Self-alignment (or initial alignment) is the process by which the Inertial Navigation System (INS) is aligned using only measurements from the inertial sensors and the reference navigation information in the stationary state. The main purpose of self-alignment is to calculate the initial attitude of the INS. The accuracy of self-alignment is determined by the performance grade of the inertial sensors, for instance, the accuracy of the horizontal attitude by the horizontal accelerometer and the accuracy of the vertical attitude by the East-axis gyro. Therefore, uncertain errors in the inertial sensors degrade the performance of self-alignment. The focus of this paper is the temperature stabilizing error of accelerometers, a form of uncertain error. An analysis is presented of how the temperature stabilizing error affect the accuracy of self-alignment. From the analysis, a method is proposed to improve performance by curve fitting the horizontal control rates. This is then verified experimentally.

Accurate Measurement Calculation Method for Interferometric Radar Altimeter-Based Terrain Referenced Navigation.

In order to improve the performance of Terrain Referenced Navigation (TRN), an Interferometric Radar Altimeter (IRA) has been developed as a more accurate altimeter. The IRA outputs not only the relative distance (slant range, R) but also the cross-track angle (look angle, θ) of the closest point on the zero Doppler line by using the principle of interferometry and two or more antennas. To perform TRN using the IRA, the 3D relative position of the closest point should be calculated. There is a formula to calculate the relative position of the closest point using the Euler angles. However, in an actual flight environment in which the influence of wind exists, the angle of attack, the side slip angle and "the effective look angle" should be used rather than the Euler angles. In this paper, a new formula for calculating the relative position of the closest point is proposed and mathematically derived. The proposed formula was verified with real data from actual flight. The flight test results show that the positions of the closest points calculated using the conventional method and the proposed method are different because of the wind effect. The TRN simulation results indicate that the proposed formula calculates the closest points more accurately than the conventional formula.

Real-time augmented model guidance for mandibular proximal segment repositioning in orthognathic surgery, using electromagnetic tracking.

It is essential to reposition the mandibular proximal segment (MPS) as close to its original position as possible during orthognathic surgery. Conventional methods cannot pinpoint the exact position of the condyle in the fossa in real time during repositioning. In this study, based on an improved registration method and a separable electromagnetic tracking tool, we developed a real-time, augmented, model-guided method for MPS surgery to reposition the condyle into its original position more accurately. After virtual surgery planning, using a complex maxillomandibular model, the final position of the virtual MPS model was simulated via 3D rotations. The displacements resulting from the MPS simulation were applied to the MPS landmarks to indicate their final postoperative positions. We designed a new registration body with 24 fiducial points for registration, and determined the optimal point group on the registration body through a phantom study. The registration between the patient's CT image and physical spaces was performed preoperatively using the optimal points. We also developed a separable frame for installing the electromagnetic tracking tool on the patient's MPS. During MPS surgery, the electromagnetic tracking tool was repeatedly attached to, and separated from, the MPS using the separable frame. The MPS movement resulting from the surgeon's manipulation was tracked by the electromagnetic tracking system. The augmented condyle model and its landmarks were visualized continuously in real time with respect to the simulated model and landmarks. Our method also provides augmented 3D coronal and sagittal views of the fossa and condyle, to allow the surgeon to examine the 3D condyle-fossa positional relationship more accurately. The root mean square differences between the simulated and intraoperative MPS models, and between the simulated and postoperative CT models, were 1.71 ± 0.63 mm and 1.89 ± 0.22 mm respectively at three condylar landmarks. Thus, the surgeons could perform MPS repositioning conveniently and accurately based on real-time augmented model guidance on the 3D condyle positional relationship with respect to the glenoid fossa, using augmented and simulated models and landmarks.

Design of Enhanced Rotation Locked Loop for Roll Angle Estimation of Rotating Vehicle in a Weak GPS Signal Environment.

In order to estimate the roll angle of a rotating vehicle, an enhanced rotation locked loop (RLL) algorithm is proposed in this paper. The RLL algorithm estimates the roll angle by using the property that the power of the GPS signal measured at the receiver of a rotating vehicle changes periodically. However, in case the received GPS power is decreased, the performance of the conventional RLL algorithm degrades, or it cannot estimate the roll angle anymore, therefore, for operating the RLL algorithm in a weak signal environment, this paper designs a method to increase the signal-to-noise ratio (SNR) by overlapping multiple GPS signals' correlator outputs and a method to compensate the decreased response of a rotation discriminator at low-signal strength. Through computer simulations, the performance of the proposed algorithm is verified and it is shown that the roll angle can be estimated stably even at a weak signal environment down to 29 dB⁻Hz of C/N0.

Autonomous bone reposition around anatomical landmark for robot-assisted orthognathic surgery.

The purpose of this study was to develop a new method for enabling a robot to assist a surgeon in repositioning a bone segment to accurately transfer a preoperative virtual plan into the intraoperative phase in orthognathic surgery. We developed a robot system consisting of an arm with six degrees of freedom, a robot motion-controller, and a PC. An end-effector at the end of the robot arm transferred the movements of the robot arm to the patient's jawbone. The registration between the robot and CT image spaces was performed completely preoperatively, and the intraoperative registration could be finished using only position changes of the tracking tools at the robot end-effector and the patient's splint. The phantom's maxillomandibular complex (MMC) connected to the robot's end-effector was repositioned autonomously by the robot movements around an anatomical landmark of interest based on the tool center point (TCP) principle. The robot repositioned the MMC around the TCP of the incisor of the maxilla and the pogonion of the mandible following plans for real orthognathic patients. The accuracy of the robot's repositioning increased when an anatomical landmark for the TCP was close to the registration fiducials. In spite of this influence, we could increase the repositioning accuracy at the landmark by using the landmark itself as the TCP. With its ability to incorporate virtual planning using a CT image and autonomously execute the plan around an anatomical landmark of interest, the robot could help surgeons reposition bones more accurately and dexterously.

Removal of copper, nickel and chromium mixtures from metal plating wastewater by adsorption with modified carbon foam.

In this study, the characterizations and adsorption efficiencies for chromium, copper and nickel were evaluated using manufacture-grade Fe2O3-carbon foam. SEM, XRD, XRF and BET analyses were performed to determine the characteristics of the material. Various pore sizes (12-420 µm) and iron contents (3.62%) were found on the surface of the Fe2O3-carbon foam. Fe2O3-carbon foam was found to have excellent adsorption efficiency compared to carbon foam for mixed solutions of cationic and anionic heavy metals. The adsorption capacities for chromium, copper and nickel were 6.7, 3.8 and 6.4 mg/g, respectively, which were obtained using a dual exponential adsorption model. In experiments with varying dosages of the Fe2O3 powder, no notable differences were observed in the removal efficiency. In a fixed-bed column test, Fe2O3-carbon foam achieved adsorption capacities for chromium, copper and nickel of 33.0, 12.0 and 9.5 mg/g, respectively, after 104 h. Based on these results, Fe2O3-carbon foam was observed to be a promising material for treatment of plating wastewater.

Virtual skeletal complex model- and landmark-guided orthognathic surgery system.

In this study, correction of the maxillofacial deformities was performed by repositioning bone segments to an appropriate location according to the preoperative planning in orthognathic surgery. The surgery was planned using the patient's virtual skeletal models fused with optically scanned three-dimensional dentition. The virtual maxillomandibular complex (MMC) model of the patient's final occlusal relationship was generated by fusion of the maxillary and mandibular models with scanned occlusion. The final position of the MMC was simulated preoperatively by planning and was used as a goal model for guidance. During surgery, the intraoperative registration was finished immediately using only software processing. For accurate repositioning, the intraoperative MMC model was visualized on the monitor with respect to the simulated MMC model, and the intraoperative positions of multiple landmarks were also visualized on the MMC surface model. The deviation errors between the intraoperative and the final positions of each landmark were visualized quantitatively. As a result, the surgeon could easily recognize the three-dimensional deviation of the intraoperative MMC state from the final goal model without manually applying a pointing tool, and could also quickly determine the amount and direction of further MMC movements needed to reach the goal position. The surgeon could also perform various osteotomies and remove bone interference conveniently, as the maxillary tracking tool could be separated from the MMC. The root mean square (RMS) difference between the preoperative planning and the intraoperative guidance was 1.16 ± 0.34 mm immediately after repositioning. After surgery, the RMS differences between the planning and the postoperative computed tomographic model were 1.31 ± 0.28 mm and 1.74 ± 0.73 mm for the maxillary and mandibular landmarks, respectively. Our method provides accurate and flexible guidance for bimaxillary orthognathic surgery based on intraoperative visualization and quantification of deviations for simulated postoperative MMC and landmarks. The guidance using simulated skeletal models and landmarks can complement and improve conventional navigational surgery for bone repositioning in the craniomaxillofacial area.

Bioabsorbable bone plates enabled with local, sustained delivery of alendronate for bone regeneration.

We prepared a bone plate enabled with the local, sustained release of alendronate, which is a drug known to inhibit osteoclast-mediated bone resorption and also expedite the bone-remodeling activity of osteoblasts. For this, we coated a bone plate already in clinical use (PLT-1031, Inion, Finland) with a blend of alendronate and a biocompatible polymer, azidobenzoic acid-modified chitosan (i.e., Az-CH) photo-crosslinked by UV irradiation. As we performed the in vitro drug release study, the drug was released from the coating at an average rate of 4.03µg/day for 63days in a sustained manner. To examine the effect on bone regeneration, the plate was fixed on an 8mm cranial critical size defect in living rats and the newly formed bone volume was quantitatively evaluated by micro-computed tomography (micro-CT) at scheduled times over 8weeks. At week 8, the group implanted with the plate enabled with sustained delivery of alendronate showed a significantly higher volume of newly formed bone (52.78±6.84%) than the groups implanted with the plates without drug (23.6±3.81%) (p<0.05). The plate enabled with alendronate delivery also exhibited good biocompatibility on H&E staining, which was comparable to the Inion plate already in clinical use. Therefore, we suggest that a bone plate enabled with local, sustained delivery of alendronate can be a promising system with the combined functionality of bone fixation and its expedited repair.

Immunogenicity and safety of pneumococcal 7-valent conjugate vaccine (diphtheria CRM(197) protein conjugate; Prevenar ) in Korean infants: differences that are found in Asian children.

This study was conducted to determine the immunogenicity and safety of a 7-valent CRM197 protein conjugated pneumococcal vaccine (PCV7) in Korean infants immunized at 2, 4 and 6 months. A total of 202 infants were enrolled and 146 and 141 infants were, respectively, included in post-2nd dose and post-3rd dose immunogenicity evaluations conducted on a per protocol basis. After two and three PCV7 vaccinations, 63.0-98.0 and 97.2-100% of infants achieved an antibody level of >or=0.35microg/mL, respectively, with a lowest against serotype 6B. No vaccination-related serious adverse reactions were observed. Thus, PCV7 appears safe and highly immunogenic in Korean infants, and adopting two doses for a primary series could be a feasible option for facilitating vaccine coverage rate.