Landmark Positions Research Articles

Assessing 3D volumetric asymmetry in facial palsy patients via advanced multi-view landmarks and radial curves

The research on facial palsy, a unilateral palsy of the facial nerve, is a complex field with many different causes and symptoms. Even modern approaches to evaluate the facial palsy state rely mainly on stills and 2D videos of the face and rarely on dynamic 3D information. Many of these analysis and visualization methods require manual intervention, which is time-consuming and error-prone. Moreover, they often depend on alignment algorithms or Euclidean measurements and consider only static facial expressions. Volumetric changes by muscle movement are essential for facial palsy analysis but require manual extraction. We propose to extract an estimated unilateral volumetric description for dynamic expressions from 3D scans. Accurate landmark positioning is required for processing the unstructured facial scans. In our case, it is attained via a multi-view method compatible with any existing 2D predictors. We analyze prediction stability and robustness against head rotation during video sequences. Further, we investigate volume changes in static and dynamic facial expressions for 34 patients with unilateral facial palsy and visualize volumetric disparities on the face surface. In a case study, we observe a decrease in the volumetric difference between the face sides during happy expressions at the beginning (13.8 ± 10.0 mm3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {mm}^{3}$$\\end{document}) and end (12.8 ± 10.3 mm3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {mm}^{3}$$\\end{document}) of a ten-day biofeedback therapy. The neutral face kept a consistent volume range of 11.8-\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$-$$\\end{document}12.1 mm3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {mm}^3$$\\end{document}. The reduced volumetric difference after therapy indicates less facial asymmetry during movement, which can be used to monitor and guide treatment decisions. Our approach minimizes human intervention, simplifying the clinical routine and interaction with 3D scans to provide a more comprehensive analysis of facial palsy.

Sphaeroptica: A tool for pseudo-3D visualization and 3D measurements on arthropods.

Natural history collections are invaluable reference collections. Digitizing these collections is a transformative process that improves the accessibility, preservation, and exploitation of specimens and associated data in the long term. Arthropods make up the majority of zoological collections. However, arthropods are small, have detailed color textures and share small, complex and shiny structures, which poses a challenge to conventional digitization methods. Sphaeroptica is a multi-images viewer that uses a sphere of oriented images. It allows the visualization of insects including their tiniest features, the positioning of landmarks, and the extraction of 3D coordinates for measuring linear distances or for use in geometric morphometrics analysis. The quantitative comparisons show that the measures obtained with Sphaeroptica are similar to the measurements derived from 3D μCT models with an average difference inferior to 1%, while featuring the high resolution of color stacked pictures with all details like setae, chaetae, scales, and other small and/or complex structures. Shaeroptica was developed for the digitization of small arthropods but it can be used with any sphere of aligned images resulting from the digitization of objects or specimens with complex surface and shining, black, or translucent texture which cannot easily be digitized using structured light scanner or Structure-from-Motion (SfM) photogrammetry.

Research progress on femoral attachment positioning during medial patellofemoral ligament reconstruction

To review research progress on femoral attachment positioning during medial patellofemoral ligament (MPFL) reconstruction, so as to provide a reference for accurate positioning in clinic. The literature at home and abroad on femoral attachment positioning during MPFL reconstruction was extensively reviewed and summarized. MPFL is the main ligament that restricts patellar outward migration, so MPFL reconstruction is the main treatment for patellar dislocation, but the accuracy of intraoperative femoral attachment positioning will significantly affect the effectiveness. At present, there are three main methods for femoral attachment positioning in MPFL reconstruction, including imaging positioning, bony landmark positioning, and new technology. Among them, the main imaging positioning method is the "Schöttle point" method, but it has high requirements for fluoroscopic positioning, and can only be accurately positioned under standard lateral fluoroscopy of the femur. The bony landmark positioning method mainly locates the femoral attachment by touching or dissecting the bony landmarks such as adductor tubercles and medial epicondyle of femur, but its disadvantages are that the positioning is not accurate enough, the intraoperative visual field exposure requirements are high, and a large incision is required. In order to avoid the problem that the simple bony landmark positioning method, in recent years, the combination of bony landmarks combined with arthroscopy, three-dimensional (3D) printing technology, and robot-assisted positioning methods have begun to be used in clinical practice. New technology localization methods have shown good results by preparing guides before operation, planning positioning paths in advance, or directly using robots to assist positioning during operation. The accurate positioning of the femoral attachment in MPFL reconstruction is crucial, and the method of accurate and rapid intraoperative determination needs to be further improved and optimized. In the future, it is expected that the combination of computer image recognition correction technology and intraoperative position assistance will solve this problem.

Evaluation of the RF depositions at 3T in routine clinical scans with respect to the SAR safety to improve efficiency of MRI utilization.

This study explores the potential for improving of 3T MRI utilization by assessing and tailoring RF exposure in routine clinical scans while complying to standard safety limit. Using two generic human body models, we evaluated pbSAR10g values at four landmark positions (knee, pelvis, thoracic spine, head) at different wbSAR levels. Specifically, weanalyzed local SAR10g in different operating modes and computed the maximum safety wbSAR, ensuring compliance with IEC limits. In normal operating mode, the RF power deposition reached wbSAR limit before the pbSAR10g limit. In the first level controlled operating mode, pbSAR10g limit is reached before the wbSAR limit in the knee, thoracic spine scanning scenarios, while the wbSAR limit is reached first in the pelvis scanning scenarios, making it the most potential-releasing (up to 33.33 %) scanning scenario. For head exposure, the head SAR10g limit is reached before the wbSAR limit, highlighting the necessity for strict SAR control. Moreover, we calculated the minimum allowable TR for common imaging sequences for reference. Different RF exposure setups are necessary to meet safety standards in various scenarios. By implementing careful RF exposure setups in routine clinical scans, the high potential capacity of 3T MRI can be fully released.

Photogrammetry is a useful tool to assess the aesthetic outcome after excision and reconstruction of the nose skin tumors

BackgroundobjectivePost-oncological nasal reconstruction presents both aesthetic and functional challenges. While established methods exist for quantitatively evaluating functional results following surgery, equivalent systems for assessing aesthetic outcomes are lacking. Three-dimensional (3D) photogrammetry, already used in maxillofacial and orthodontic surgery for aesthetic evaluation, overcomes some limitations of traditional methods like direct anthropometry. However, its applicability in oncological facial reconstruction has not yet been explored. In our study, we applied the 3dMDtrio™ system for the quantitative analysis of line and surface modifications following nasal reconstruction. MethodsWe conducted a prospective observational study enrolling patients with skin neoplasms located on the nose undergoing surgical excision and reconstruction. Using the 3dMDtrio™ system, we measured the dimensions and projections of nasal surfaces and the positions of specific landmarks before and after surgery. The surface measurements were then correlated with aesthetic evaluations performed by three plastic surgeons, not involved in the procedure, using a 5-point Likert scale. ResultsWe included 33 patients with a mean age of 71 years, ranging from 40 to 94. We obtained complete documentation of all postoperative measurements for 21 patients. We observed significant changes in the positions of the landmarks post-surgery, limited to the right ala and nasion. The average nasal surface area was 4674.41 mm2 ± 477.24 mm2 before surgery and 4667.95 mm2 ± 474.12 mm2 after surgery, with no significant discrepancies. The evaluation using the Likert scale revealed an average score of 3.04 ± 0.48, with a significant negative correlation to the measured surface changes. ConclusionOur findings suggest that 3D photogrammetry can be considered a valid method for objectively assessing volumetric changes associated with post-oncological nasal reconstructive surgery.

Landmark Positioning on a Map: An Alternative Measure of Spatial Ability for Identifying Students Who May Benefit From Learning Gross Anatomy with Virtual Reality.

Research has shown an inconsistent relationship between spatial abilities and learning outcomes from virtual anatomical tools. Instructors must understand this relationship to select appropriate resources for diverse learners. To identify appropriate tests for measuring spatial ability and evaluate the effectiveness of virtual anatomical resources, this study compared 96 students' visuospatial ability (measured using the Mental Rotation Task [MRT] and Landmark Position on a Map [LPM] tests) with learning outcomes from experimental anatomy sessions and undergraduate anatomical course examinations. During experimental sessions, students took a test after a brief instructional session using one virtual resource: a monoscopic resource (e.g., digital photographs or a rotatable three-dimensional [r3D] specimen) or a stereoscopic virtual reality (VR) specimen. A negative linear relationship was found between MRT scores and students in Session B using VR with controllers (r= -.56 to -.29), and LPM scores and students using VR (r= -.71 to .39) and r3D (r= -.41 to .43). There was a positive linear relationship between MRT scores and all other resources (r= .01 to .91), and course examination scores (r= .25 to .42, p = .05). Although the results were inconsistent, correlations were found between spatial ability and outcomes using both the MRT and LPM. The LPM might be better suited for determining which learners would benefit from VR. The results suggest that monoscopic resources best support high spatial abilities, and stereoscopic resources best support low spatial abilities. These findings support accounting for diverse learner visuospatial abilities when selecting resources.

Automatic localization of anatomical landmarks in head cine fluoroscopy images via deep learning.

Fluoroscopy guided interventions (FGIs) pose a risk of prolonged radiation exposure; personalized patient dosimetry is necessary to improve patient safety during these procedures. However, current FGIs systems do not capture the precise exposure regions of the patient, making it challenging to perform patient-procedure-specific dosimetry. Thus, there is a pressing need to develop approaches to extract and use this information to enable personalized radiation dosimetry for interventional procedures. To propose a deep learning (DL) approach for the automatic localization of 3D anatomical landmarks on randomly collimated and magnified 2D head fluoroscopy images. The model was developed with datasets comprising 800000 pseudo 2D synthetic images (mixture of vessel-enhanced and non-enhancement), each with 55 annotated anatomical landmarks (two are landmarks for eye lenses), generated from 135 retrospectively collected head computed tomography (CT) volumetric data. Before training, dynamic random cropping was performed to mimic the varied field-size collimation in FGI procedures. Gaussian-distributed additive noise was applied to each individual image to enhance the robustness of the DL model in handling image degradation that may occur during clinical image acquisition in a clinical environment. The model was trained with 629370 synthetic images for approximately 275000 iterations and evaluated against a synthetic image test set and a clinical fluoroscopy test set. The model shows good performance in estimating in- and out-of-image landmark positions and shows feasibility to instantiate the skull shape. The model successfully detected 96.4% and 92.5% 2D and 3D landmarks, respectively, within a 10mm error on synthetic test images. It demonstrated an average of 3.6±2.3mm mean radial error and successfully detected 96.8% 2D landmarks within 10mm error on clinical fluoroscopy images. Our deep-learning model successfully localizes anatomical landmarks and estimates the gross shape of skull structures from collimated 2D projection views. This method may help identify the exposure region required for patient-specific organ dosimetry in FGIs procedures.

Heatmap-Based Active Shape Model for Landmark Detection in Lumbar X-ray Images.

Medical staff inspect lumbar X-ray images to diagnose lumbar spine diseases, and the analysis process is currently automated using deep-learning techniques. The detection of landmarks is necessary in the automatic process of localizing the position and identifying the morphological features of the vertebrae. However, detection errors may occur owing to the noise and ambiguity of images, as well as individual variations in the shape of the lumbar vertebrae. This study proposes a method to improve the robustness of landmark detection results. This method assumes that landmarks are detected by a convolutional neural network-based two-step model consisting of Pose-Net and M-Net. The model generates a heatmap response to indicate the probable landmark positions. The proposed method then corrects the landmark positions using the heatmap response and active shape model, which employs statistical information on the landmark distribution. Experiments were conducted using 3600 lumbar X-ray images, and the results showed that the landmark detection error was reduced by the proposed method. The average value of maximum errors decreased by 5.58% after applying the proposed method, which combines the outstanding image analysis capabilities of deep learning with statistical shape constraints on landmark distribution. The proposed method could also be easily integrated with other techniques to increase the robustness of landmark detection results such as CoordConv layers and non-directional part affinity field. This resulted in a further enhancement in the landmark detection performance. These advantages can improve the reliability of automatic systems used to inspect lumbar X-ray images. This will benefit both patients and medical staff by reducing medical expenses and increasing diagnostic efficiency.

Isotopological remeshing and statistical shape analysis: Enhancing premolar tooth wear classification and simulation with machine learning

ObjectiveThe aim of this study was to evaluate the accuracy of a combined approach based on an isotopological remeshing and statistical shape analysis (SSA) to capture key anatomical features of altered and intact premolars. Additionally, the study compares the capabilities of four Machine Learning (ML) algorithms in identifying or simulating tooth alterations. Methods113 premolar surfaces from a multicenter database were analyzed. These surfaces were processed using an isotopological remeshing method, followed by a SSA. Mean Euclidean distances between the initial and remeshed STL files were calculated to assess deviation in anatomical landmark positioning. Seven anatomical features were extracted from each tooth, and their correlations with shape modes and morphological characteristics were explored. Four ML algorithms, validated through three-fold cross-validation, were assessed for their ability to classify tooth types and alterations. Additionally, twenty intact teeth were altered and then reconstructed to verify the method's accuracy. ResultsThe first five modes encapsulated 76.1% of the total shape variability, with a mean landmark positioning deviation of 10.4 µm (±6.4). Significant correlations were found between shape modes and specific morphological features. The optimal ML algorithms demonstrated high accuracy (>83%) and precision (>86%). Simulations on intact teeth showed discrepancies in anatomical features below 3%. ConclusionThe combination of an isotopological remeshing with SSA showed good reliability in capturing key anatomical features of the tooth. Clinical significanceThe encouraging performance of ML algorithms suggests a promising direction for supporting practitioners in diagnosing and planning treatments for patients with altered teeth, ultimately improving preventive care.

The impact of transcranial direct current stimulation combined with interim testing on spatial route learning in patients with schizophrenia

BackgroundCognitive deficits in patients with schizophrenia have drawn widespread attention. Transcranial direct current stimulation (tDCS) can modulate cognitive processes by altering neuronal excitability. Previous studies have found that interim testing can enhance spatial route learning and memory in patients with schizophrenia. However, there has been limited research on the combined effects of these two methods on spatial route learning in these patients. ObjectiveTo investigate whether the combination of tDCS and interim testing can effectively contribute to the maintenance of spatial route memory in patients with schizophrenia. The study involved conducting route learning using interim testing after anodal tDCS treatment on the left dorsolateral prefrontal cortex (L-DLPFC). MethodsNinety-two patients with schizophrenia were recruited and divided into groups receiving anodal, sham, or no stimulation. The anodal group received L-DLPFC tDCS treatment 10 times over 5 days (twice daily for 20 min). After treatment, spatial route learning was assessed in interim testing. Correct recall rates of landmark positions and proactive interference from prior learning were compared among the groups. ResultsRegardless of stimulation type, the interim testing group outperformed the relearning group. Additionally, recall scores were higher following anodal stimulation, indicating the efficacy of tDCS. ConclusionsBoth tDCS and interim testing independently enhance the ability to learn new information in spatial route learning for patients with schizophrenia, indicating that tDCS of the left DLPFC significantly improves memory in these patients.

Poster 383: The Orthopaedic Vital Sign: Feasibility and Capability of Smartphone-Based Kinematic Analysis in Clinic for Functional Movement Evaluation

Objectives: Rehabilitation progression and return-to-play (RTP) guidelines vary greatly and are predominantly based on expert opinion and subjective criteria. Evaluating joint acceleration and forces in a biomechanical laboratory setting require substantial resources and are not feasible within a clinical setting for a multitude of patients. Video motion capture systems on smartphone devices offer a modern and implementable biomechanical assessment capability. Our purpose is to (1) show the in-clinic practicality of an AI-driven smartphone motion capture system for analyzing human motion, (2) identify movements and biomechanical parameters best correlating with clinical evaluation, and lastly predicting rehabilitation progression and return-to-play. We hypothesize that the smartphone motion capture system OpenCap is easy implementable in clinic and can identify biomechanical movement parameters which will streamline rehabilitation progression and RTP following various sports injuries and surgeries. This study is part of the first phase of the financially supported AOSSM Playmaker Grant research project. Methods: Healthy controls and participants with various musculoskeletal conditions (e.g. nonoperative treated anterior knee pain, post ACL reconstruction, knee osteoarthritis) were recruited. Participants performed defined movements that were captured with the smartphone-based 3-camera video motion capture system OpenCap in the outpatient clinical setting. OpenCap is a smartphone-based open-source software for estimating the kinematics and kinetics of human movement. Patients were grouped into three categories, easy, intermediate, and hard, based on the quality of performed movements, as deemed by the research team, the patients’ self-assessments, and restrictions set by the treating physician. The advanced groups completed quantitatively more and qualitatively higher joint loading movements. All movements were performed three times and during analysis averaged. The easy testbattery consisted of the timed up-and-go test, five-second slow sit-to-stand test and a 10 second one-leg lift for balance. The intermediate group additionally performed 2-legged squats and 1-legged pistol squats. The hard group performed on top also single leg Romanian deadlifts, vertical hop tests for height, step-down tests and drop jumps. OpenCap recorded the movement and computed positions of anatomic landmarks, joint angles, and joint forces during the movements (Figure 1). The time of motion capture system setup and movement completion were recorded. Various biomechanical parameters were explored and compared between groups to identify the best biomechanical parameters and movements allowing group distinguishment. Results: 100 participants visiting a tertiary orthopaedic surgery clinic were included, consisting of 20 healthy controls and 80 patients with various nonoperative and surgical treated pathologies. The cohort consists of 51% females and has an average age of 41.4 years. All participants completed the easy test battery, 53% the intermediate test battery, and 22% the hard test battery. The average setup time was 5.3 minutes, including acquiring patients’ age, height, and weight necessary for the musculoskeletal model and calibration, but excluding obtaining an informed consent. The movement capturing time differed between testbattery groups and ranged from 3 to 17 minutes (average easy 3.6 min, intermediate 5.2 min, hard 11.3 min). Setup time was reduced during the continuation of the study from 9.2 min in the first 20 participants to 3.4min in the last 20 participants. No injuries occurred. In our exploratory analysis, hip flexion, hip adduction, and hip rotation most clearly displayed differences between the easy and intermediate classified groups from the hard motion participants. Comparing the minimum hip flexion between the groups during the slow sit-to-stand test, the easy and intermediate group showed a 5° increased hip flexion (easy p=0.000, intermediate p=0.004) compared to the hard group. During the single leg squat the maximum hip flexion was 15° less in the intermediate compared to the hard group (p=0.009). During the lateral step-down test, the intermediate group showed a minimum 17° decreased hip internal rotation in comparison to the hard group (p=0.015). The slow sit to stand test, timed up and go test, one leg lift, single leg squat and the lateral step-down test demonstrated best discrimination between the groups. Conclusions: The video motion capture system OpenCap can be easily implemented into a clinical setting due to the easy setup and expeditious manner in which the measurements are taken. Testing time is mostly dependent on the extent of measured movements. Moreover, biomechanical parameters as assessed by OpenCap allow differentiation between physical capability groups, compared to clinically assessed. The single-legged movements, the slow sit-to-stand test and the timed-up-and-go-test are most effective in revealing these biomechanical differences. The biomechanical parameters of the hip, including its flexion, adduction, and rotation, reveal these differences most clearly. The ultimate goal is to provide a clinically friendly, rapid, inexpensive, portable, and objective assessment tool to facilitate rehabilitation and return to sport following various injuries and surgeries.

Accurate prediction of neurologic changes in critically ill infants using pose AI.

Infant alertness and neurologic changes can reflect life-threatening pathology but are assessed by exam, which can be intermittent and subjective. Reliable, continuous methods are needed. We hypothesized that our computer vision method to track movement, pose AI, could predict neurologic changes in the neonatal intensive care unit (NICU). We collected 4,705 hours of video linked to electroencephalograms (EEG) from 115 infants. We trained a deep learning pose algorithm that accurately predicted anatomic landmarks in three evaluation sets (ROC-AUCs 0.83-0.94), showing feasibility of applying pose AI in an ICU. We then trained classifiers on landmarks from pose AI and observed high performance for sedation (ROC-AUCs 0.87-0.91) and cerebral dysfunction (ROC-AUCs 0.76-0.91), demonstrating that an EEG diagnosis can be predicted from video data alone. Taken together, deep learning with pose AI may offer a scalable, minimally invasive method for neuro-telemetry in the NICU.

Strategies for Minimizing the Effects of Observer Variability on Sagittal Parameter Measurements of the Spine.

Exploratory-descriptive study. Evaluate the influence of observer variability on the measurements of both thoracic kyphosis (TK) and lumbar lordosis (LL) obtained with anatomic and functional spinal segmentation methods. Parametric analysis for spinal surgery planning typically relies on anatomic parameters. However, incorporating functional parameters that consider the vertebrae orientation is important to minimizing surgical calculation errors. The authors developed parametric analysis software that integrates traditional and functional methodologies. The proposed method included functional thoracic kyphosis and functional lumbar lordosis calculated from the lines normal to the inflection points of the spine model. Using a synthetic lateral X-ray, the observer variability was computer-simulated generating 20 landmark sets that replicate the annotations of 20 observers. The analysis also included 10 clinical X-rays, annotated twice by 3 judges with a minimum 1-week interval. The spinal curvature angles were derived using the anatomic and functional methods. Statistical analysis were performed for comparison. For the synthetic X-ray, the proposed method presented significantly less variability: TK (<±2.5 degrees, P=0.00023) and LL (<±5 degrees, P=0.00012). For the clinical X-rays, the interobserver reliability analysis yielded higher intraclass correlation coefficients (ICC) for functional TK (ICC>0.97) and functional LL (ICC>0.87) than for TK (ICC<0.91) and LL (ICC<0.89). Statistically significant differences were observed for both TK (P=0.001) and LL (P=0.030). Under the traditional method, observer variability led to measurement differences surpassing ±19 degrees, whereas differences with the proposed method were within ±10 degrees for both parameters. The vertebral endplate is not the most suitable place to measure spinal sagittal curvatures. Small changes in landmark position significantly alter the measured Cobb angle. The proposed method offers a substantial advantage regarding the influence of observer variability, in addition to the more individualized analysis.

The Via Severiana and the Tabula Peutingeriana: valuation of landmarks precision and town expansions

This work deals with the examination of the route of the Via Severiana, traced in the imperial period, from a historical and technical point of view. Then, results were compared with the Tabula, which we can consider one of the first synthetic representations of general viability. We have explored the development of the various sectors of the via Severiana, considering the various utility and presence on the territory, using the Tabula Peutingeriana as a benchmark. Subsequently, we have examined the precision of the positioning of the various landmarks along the via Severiana and the movement of the inhabited centers as a function of density. Modern technologies, including Geographic Information Systems and mathematical models, allow us to help the archaeologists to overlay ancient maps like the Tabula Peutingeriana onto contemporary maps, aiding in the identification of locations and understanding ancient landscapes.

Experimental validation of a PNS-optimized whole-body gradient coil.

Peripheral nerve stimulation (PNS) limits the usability of state-of-the-art whole-body and head-only MRI gradient coils. We used detailed electromagnetic and neurodynamic modeling to set an explicit PNS constraint during the design of a whole-body gradient coil and constructed it to compare the predicted and experimentally measured PNS thresholds to those of a matched design without PNS constraints. We designed, constructed, and tested two actively shielded whole-body Y-axis gradient coil winding patterns: YG1 is a conventional symmetric design without PNS-optimization, whereas YG2's design used an additional constraint on the allowable PNS threshold in the head-imaging landmark, yielding an asymmetric winding pattern. We measured PNS thresholds in 18 healthy subjects at five landmark positions (head, cardiac, abdominal, pelvic, and knee). The PNS-optimized design YG2 achieved 46% higher average experimental thresholds for a head-imaging landmark than YG1 while incurring a 15% inductance penalty. For cardiac, pelvic, and knee imaging landmarks, the PNS thresholds increased between +22% and +35%. For abdominal imaging, PNS thresholds did not change significantly between YG1 and YG2 (-3.6%). The agreement between predicted and experimental PNS thresholds was within 11.4% normalized root mean square error for both coils and all landmarks. The PNS model also produced plausible predictions of the stimulation sites when compared to the sites of perception reported by the subjects. The PNS-optimization improved the PNS thresholds for the target scan landmark as well as most other studied landmarks, potentially yielding a significant improvement in image encoding performance that can be safely used in humans.

Automated Landmark Annotation for Morphometric Analysis of Distal Femur and Proximal Tibia.

Manual anatomical landmarking for morphometric knee bone characterization in orthopedics is highly time-consuming and shows high operator variability. Therefore, automation could be a substantial improvement for diagnostics and personalized treatments relying on landmark-based methods. Applications include implant sizing and planning, meniscal allograft sizing, and morphological risk factor assessment. For twenty MRI-based 3D bone and cartilage models, anatomical landmarks were manually applied by three experts, and morphometric measurements for 3D characterization of the distal femur and proximal tibia were calculated from all observations. One expert performed the landmark annotations three times. Intra- and inter-observer variations were assessed for landmark position and measurements. The mean of the three expert annotations served as the ground truth. Next, automated landmark annotation was performed by elastic deformation of a template shape, followed by landmark optimization at extreme positions (highest/lowest/most medial/lateral point). The results of our automated annotation method were compared with ground truth, and percentages of landmarks and measurements adhering to different tolerances were calculated. Reliability was evaluated by the intraclass correlation coefficient (ICC). For the manual annotations, the inter-observer absolute difference was 1.53 ± 1.22 mm (mean ± SD) for the landmark positions and 0.56 ± 0.55 mm (mean ± SD) for the morphometric measurements. Automated versus manual landmark extraction differed by an average of 2.05 mm. The automated measurements demonstrated an absolute difference of 0.78 ± 0.60 mm (mean ± SD) from their manual counterparts. Overall, 92% of the automated landmarks were within 4 mm of the expert mean position, and 95% of all morphometric measurements were within 2 mm of the expert mean measurements. The ICC (manual versus automated) for automated morphometric measurements was between 0.926 and 1. Manual annotations required on average 18 min of operator interaction time, while automated annotations only needed 7 min of operator-independent computing time. Considering the time consumption and variability among observers, there is a clear need for a more efficient, standardized, and operator-independent algorithm. Our automated method demonstrated excellent accuracy and reliability for landmark positioning and morphometric measurements. Above all, this automated method will lead to a faster, scalable, and operator-independent morphometric analysis of the knee.

Fluoroscopy-Assisted Computer Navigation Accurately Determines Cup Position and Leg Length for Anterior Hip Arthroplasty.

Recently, fluoroscopy-assisted computer navigation has been developed to assess intraoperative cup inclination/anteversion and leg-length discrepancy (LLD) in the operating room. However, there is a relative dearth of studies investigating the accuracy of this software compared with postoperative radiographs. We prospectively enrolled 211 navigated anterior total hip arthroplasties using fluoroscopy-assisted computer navigation software. Intraoperative navigated measurements were compared with postoperative anteroposterior radiographs to assess accuracy of cup inclination/anteversion and LLD. Continuous variables were analyzed using the Student's t test, and categorical variables were analyzed using Fisher's exact test. On postoperative radiographs, 94.3% of cups (199 of 211) were positioned within the Lewinnek "safe zone," compared with 99.1% navigated intraoperatively (P=.01). Eighty-two percent of hips (174 of 211) were navigated intraoperatively to LLDs within ±2 mm; on postoperative radiographs, 65% of hips (138 of 211) had LLDs within ±2 mm (P=.0001). Intraoperatively, 100% of hips (211 of 211) were navigated to LLDs within ±5 mm; similarly, on postoperative radiographs, 98% of hips (207 of 211) had LLDs within ±5 mm (P=.12). A novel fluoroscopy-assisted computer navigation platform accurately assessed intraoperative cup position and LLD during anterior total hip arthroplasty. Careful attention to fluoroscopic technique, positioning of radiographic landmarks, and knowledge of the limitations of fluoroscopy, including parallax effect, are important concepts that surgeons should incorporate into their decision algorithm. [Orthopedics. 2024;47(4):e174-e180.].

A Deep Learning Model for Markerless Pose Estimation Based on Keypoint Augmentation: What Factors Influence Errors in Biomechanical Applications?

In biomechanics, movement is typically recorded by tracking the trajectories of anatomical landmarks previously marked using passive instrumentation, which entails several inconveniences. To overcome these disadvantages, researchers are exploring different markerless methods, such as pose estimation networks, to capture movement with equivalent accuracy to marker-based photogrammetry. However, pose estimation models usually only provide joint centers, which are incomplete data for calculating joint angles in all anatomical axes. Recently, marker augmentation models based on deep learning have emerged. These models transform pose estimation data into complete anatomical data. Building on this concept, this study presents three marker augmentation models of varying complexity that were compared to a photogrammetry system. The errors in anatomical landmark positions and the derived joint angles were calculated, and a statistical analysis of the errors was performed to identify the factors that most influence their magnitude. The proposed Transformer model improved upon the errors reported in the literature, yielding position errors of less than 1.5 cm for anatomical landmarks and 4.4 degrees for all seven movements evaluated. Anthropometric data did not influence the errors, while anatomical landmarks and movement influenced position errors, and model, rotation axis, and movement influenced joint angle errors.

PELE scores: pelvic X-ray landmark detection with pelvis extraction and enhancement.

Pelvic X-ray (PXR) is widely utilized in clinical decision-making associated with the pelvis, the lower part of the trunk that supports and balances the trunk. In particular, PXR-based landmark detection facilitates downstream analysis and computer-assisted diagnosis and treatment of pelvic diseases. Although PXR has the advantages of low radiation and reduced cost compared to computed tomography (CT), it characterizes the 2D pelvis-tissue superposition of 3D structures, which may affect the accuracy of landmark detection in some cases. However, the superposition nature of PXR is implicitly handled by existing deep learning-based landmark detection methods, which mainly design the deep network structures for better detection performances. Explicit handling of the superposition nature of PXR is rarely done. In this paper, we explicitly focus on the superposition of X-ray images. Specifically, we propose a pelvis extraction (PELE) module that consists of a decomposition network, a domain adaptation network, and an enhancement module, which utilizes 3D prior anatomical knowledge in CT to guide and well isolate the pelvis from PXR, thereby eliminating the influence of soft tissue for landmark detection. The extracted pelvis image, after enhancement, is then used for landmark detection. We conduct an extensive evaluation based on two public and one private dataset, totaling 850 PXRs. The experimental results show that the proposed PELE module significantly improves the accuracy of PXRs landmark detection and achieves state-of-the-art performances in several benchmark metrics. The design of PELE module can improve the accuracy of different pelvic landmark detection baselines, which we believe is obviously conducive to the positioning and inspection of clinical landmarks and critical structures, thus better serving downstream tasks. Our project has been open-sourced at https://github.com/ECNUACRush/PELEscores .

Enhancing endoscopic measurement: validating a quantitative method for polyp size and location estimation in upper gastrointestinal endoscopy

BackgroundAccurate measurement of polyps size is crucial in predicting malignancy, planning relevant intervention strategies and surveillance schedules. Endoscopists’ visual estimations can lack precision. This study builds on our prior research, with the aim to evaluate a recently developed quantitative method to measure the polyp size and location accurately during a simulated endoscopy session.MethodsThe quantitative method merges information about endoscopic positions obtained from an electromagnetic tracking sensor, with corresponding points on the images of the segmented polyp border. This yields real-scale 3D coordinates of the border of the polyp. By utilising the sensor, positions of any anatomical landmarks are attainable, enabling the estimation of a polyp’s location relative to them. To verify the method’s reliability and accuracy, simulated endoscopies were conducted in pig stomachs, where polyps were artificially created and assessed in a test–retest manner. The polyp measurements were subsequently compared against clipper measurements.ResultsThe average size of the fifteen polyps evaluated was approximately 12 ± 4.3 mm, ranging from 5 to 20 mm. The test–retest reliability, measured by the Intraclass Correlation Coefficient (ICC) for polyp size estimation, demonstrated an absolute agreement of 0.991 (95% CI 0.973–0.997, p < 0.05). Bland & Altman analysis revealed a mean estimation difference of − 0.17 mm (− 2.03%) for polyp size and, a mean difference of − 0.4 mm (− 0.21%) for polyp location. Both differences were statistically non-significant (p > 0.05). When comparing the proposed method with calliper measurements, the Bland & Altman plots showed 95% of size estimation differences between − 1.4 and 1.8 mm (− 13 to 17.4%) which was not significant (p > 0.05).ConclusionsThe proposed method of measurements of polyp size and location was found to be highly accurate, offering great potential for clinical implementation to improve polyp assessment. This level of performance represents a notable improvement over visual estimation technique used in clinical practice.