Self-Navigation with Compressed Sensing for 2D Translational Motion Correction in Free-Breathing Coronary MRI: A Feasibility Study

Purpose

Respiratory motion correction remains a challenge in coronary magnetic resonance imaging (MRI) and current techniques, such as navigator gating, suffer from sub-optimal scan efficiency and ease-of-use. To overcome these limitations, an image-based self-navigation technique is proposed that uses “sub-images” and compressed sensing (CS) to obtain translational motion correction in 2D. The method was preliminarily implemented as a 2D technique and tested for feasibility for targeted coronary imaging.

Methods

During a 2D segmented radial k-space data acquisition, heavily undersampled sub-images were reconstructed from the readouts collected during each cardiac cycle. These sub-images may then be used for respiratory self-navigation. Alternatively, a CS reconstruction may be used to create these sub-images, so as to partially compensate for the heavy undersampling. Both approaches were quantitatively assessed using simulations and in vivo studies, and the resulting self-navigation strategies were then compared to conventional navigator gating.

Results

Sub-images reconstructed using CS showed a lower artifact level than sub-images reconstructed without CS. As a result, the final image quality was significantly better when using CS-assisted self-navigation as opposed to the non-CS approach. Moreover, while both self-navigation techniques led to a 69% scan time reduction (as compared to navigator gating), there was no significant difference in image quality between the CS-assisted self-navigation technique and conventional navigator gating, despite the significant decrease in scan time.

Conclusions

CS-assisted self-navigation using 2D translational motion correction demonstrated feasibility of producing coronary MRA data with image quality comparable to that obtained with conventional navigator gating, and does so without the use of additional acquisitions or motion modeling, while still allowing for 100% scan efficiency and an improved ease-of-use. In conclusion, compressed sensing may become a critical adjunct for 2D translational motion correction in free-breathing cardiac imaging with high spatial resolution. An expansion to modern 3D approaches is now warranted.     

A Novel Prior- and Motion-Based Compressed Sensing Method for Small-Animal Respiratory Gated CT

Low-dose protocols for respiratory gating in cardiothoracic small-animal imaging lead to streak artifacts in the images reconstructed with a Feldkamp-Davis-Kress (FDK) method. We propose a novel prior- and motion-based reconstruction (PRIMOR) method, which improves prior-based reconstruction (PBR) by adding a penalty function that includes a model of motion. The prior image is generated as the average of all the respiratory gates, reconstructed with FDK. Motion between respiratory gates is estimated using a nonrigid registration method based on hierarchical B-splines. We compare PRIMOR with an equivalent PBR method without motion estimation using as reference the reconstruction of high dose data. From these data acquired with a micro-CT scanner, different scenarios were simulated by changing photon flux and number of projections. Methods were evaluated in terms of contrast-to-noise-ratio (CNR), mean square error (MSE), streak artefact indicator (SAI), solution error norm (SEN), and correction of respiratory motion. Also, to evaluate the effect of each method on lung studies quantification, we have computed the Jaccard similarity index of the mask obtained from segmenting each image as compared to those obtained from the high dose reconstruction. Both iterative methods greatly improved FDK reconstruction in all cases. PBR was prone to streak artifacts and presented blurring effects in bone and lung tissues when using both a low number of projections and low dose. Adopting PBR as a reference, PRIMOR increased CNR up to 33% and decreased MSE, SAI and SEN up to 20%, 4% and 13%, respectively. PRIMOR also presented better compensation for respiratory motion and higher Jaccard similarity index. In conclusion, the new method proposed for low-dose respiratory gating in small-animal scanners shows an improvement in image quality and allows a reduction of dose or a reduction of the number of projections between two and three times with respect to previous PBR approaches.     

In vivo respiratory-gated micro-CT imaging in small-animal oncology models

Micro-computed tomography(micro-CT) is becoming an accepted research tool for the noninvasive examination of laboratory animals such as mice and rats, but to date, in vivo scanning has largely been limited to the evaluation of skeletal tissues. We use a commercially available micro-CT device to perform respiratory gated in vivo acquisitions suitable for thoracic imaging. The instrument is described, along with the scan protocol and animal preparation techniques. Preliminary results confirm that lung tumors as small as 1 mm in diameter are visible in vivo with these methods. Radiation dose was evaluated using several approaches, and was found to be approximately 0.15 Gy for this respiratory-gated micro-CT imaging protocol. The combination of high-resolution CT imaging and respiratory-gated acquisitions appears well-suited to serial in vivo scanning.     

Machine learning in Magnetic Resonance Imaging: Image reconstruction

Magnetic Resonance Imaging (MRI) plays a vital role in diagnosis, management and monitoring of many diseases. However, it is an inherently slow imaging technique. Over the last 20 years, parallel imaging, temporal encoding and compressed sensing have enabled substantial speed-ups in the acquisition of MRI data, by accurately recovering missing lines of k-space data. However, clinical uptake of vastly accelerated acquisitions has been limited, in particular in compressed sensing, due to the time-consuming nature of the reconstructions and unnatural looking images. Following the success of machine learning in a wide range of imaging tasks, there has been a recent explosion in the use of machine learning in the field of MRI image reconstruction.A wide range of approaches have been proposed, which can be applied in k-space and/or image-space.Promising results have been demonstrated from a range of methods, enabling natural looking images and rapid computation.In this review article we summarize the current machine learning approaches used in MRI reconstruction, discuss their drawbacks, clinical applications, and current trends.     

Robustness of post-reconstruction and direct kinetic parameter estimates under rigid head motion in dynamic brain PET imaging

ObjectiveDynamic PET imaging is extensively used in brain imaging to estimate parametric maps. Inter-frame motion can substantially disrupt the voxel-wise time-activity curves (TACs), leading to erroneous maps during kinetic modelling. Therefore, it is important to characterize the robustness of kinetic parameters under various motion and kinetic model related factors.MethodsFully 4D brain simulations ([¹⁵O]H₂O and [¹⁸F]FDG dynamic datasets) were performed using a variety of clinically observed motion patterns. Increasing levels of head motion were investigated as well as varying temporal frames of motion initiation. Kinetic parameter estimation was performed using both post-reconstruction kinetic analysis and direct 4D image reconstruction to assess bias from inter-frame emission blurring and emission/attenuation mismatch.ResultsKinetic parameter bias heavily depends on the time point of motion initiation. Motion initiated towards the end of the scan results in the most biased parameters. For the [¹⁸F]FDG data, k₄ is the more sensitive parameter to positional changes, while K₁ and blood volume were proven to be relatively robust to motion. Direct 4D image reconstruction appeared more sensitive to changes in TACs due to motion, with parameter bias spatially propagating and depending on the level of motion.ConclusionKinetic parameter bias highly depends upon the time frame at which motion occurred, with late frame motion-induced TAC discontinuities resulting in the least accurate parameters. This is of importance during prolonged data acquisition as is often the case in neuro-receptor imaging studies. In the absence of a motion correction, use of TOF information within 4D image reconstruction could limit the error propagation.     

MR-Based PET Motion Correction Procedure for Simultaneous MR-PET Neuroimaging of Human Brain

Positron Emission Tomography (PET) images are prone to motion artefacts due to the long acquisition time of PET measurements. Recently, simultaneous magnetic resonance imaging (MRI) and PET have become available in the first generation of Hybrid MR-PET scanners. In this work, the elimination of artefacts due to head motion in PET neuroimages is achieved by a new approach utilising MR-based motion tracking in combination with PET list mode data motion correction for simultaneous MR-PET acquisitions. The method comprises accurate MR-based motion measurements, an intra-frame motion minimising and reconstruction time reducing temporal framing algorithm, and a list mode based PET reconstruction which utilises the Ordinary Poisson Algorithm and avoids axial and transaxial compression. Compared to images uncorrected for motion, an increased image quality is shown in phantom as well as in vivo images. In vivo motion corrected images show an evident increase of contrast at the basal ganglia and a good visibility of uptake in tiny structures such as superior colliculi.     

Low-Dose Micro-CT Imaging for Vascular Segmentation and Analysis Using Sparse-View Acquisitions

The aim of this study is to investigate whether reliable and accurate 3D geometrical models of the murine aortic arch can be constructed from sparse-view data in vivo micro-CT acquisitions. This would considerably reduce acquisition time and X-ray dose. In vivo contrast-enhanced micro-CT datasets were reconstructed using a conventional filtered back projection algorithm (FDK), the image space reconstruction algorithm (ISRA) and total variation regularized ISRA (ISRA-TV). The reconstructed images were then semi-automatically segmented. Segmentations of high- and low-dose protocols were compared and evaluated based on voxel classification, 3D model diameters and centerline differences. FDK reconstruction does not lead to accurate segmentation in the case of low-view acquisitions. ISRA manages accurate segmentation with 1024 or more projection views. ISRA-TV needs a minimum of 256 views. These results indicate that accurate vascular models can be obtained from micro-CT scans with 8 times less X-ray dose and acquisition time, as long as regularized iterative reconstruction is used.     

Interplane bulk motion analysis and removal based on normalized cross‐correlation in optical coherence tomography angiography

Bulk motion seriously degrades the image quality of optical coherence tomography angiography (OCTA). Conventional correction methods focus on in‐plane displacement, while the bulk motion component perpendicular to B‐scans also introduces noise. This work first presents an evaluation of this component using a specific scan protocol and an approximate expression derived from peak‐normalized cross‐correlation values, and then quantitatively assesses how interplane bulk motion noise reduce the sensitivity of cross‐sectional angiograms. Finally, we developed a repetitive bulk motion correction method based on the estimated displacements and redundant volume scans. The correction does not require registration and angiogram reconstruction of low flow sensitivity frames, and the results of in vivo mice skin OCTA imaging experiments show that the proposed method can effectively reduce bulk motion noise caused by cardiac and respiratory motion and occasional shaking, and improve OCTA image quality, which has practical significance for clinical OCTA diagnosis and analysis.     

Real-time determination of the optimal reconstruction phase to control ECG pulsing in spiral cardiac CT

The reconstruction phase providing optimal image quality in coronary CT angiography is dependent on the heart rate but additionally displays substantial patient-dependent variation. The purpose of this study was to provide online identification of the patient-specific optimal reconstruction phase during CT coronary angiography data acquisition and to allow adaptation of tube current modulation for the individual patient. A raw data-based cardiac motion signal (kymogram) was used for the detection of the optimal reconstruction phase. The individual motion curve of each patient was correlated with dedicated template curves to reduce signal noise. Data sets of 90 consecutive patients were used for validation purposes. The reliability of our approach increased with scan time and provided highest correlation with the visually identified optimal reconstruction phase already after half of the total scan time indicated by a difference value of 13.2% and 8.2%, respectively. A high correlation of the computed and the visually identified optimal reconstruction phase was assured in most cases providing a dose reduction of 36% compared to conventional TCM application for a confidence interval of 80%. Our method is a fully automatic computer-assisted approach identifying the optimal reconstruction phase with high reliability while online capability can be ensured. We conclude that our method can identify cardiac phases providing highest image quality already during CT scanning. Reduction of the tube current by a patient-specific optimization providing a minimal dose level is the major benefit for the patients.     

Optimization of cell morphology measurement via single-molecule tracking PALM

In neurons, the shape of dendritic spines relates to synapse function, which is rapidly altered during experience-dependent neural plasticity. The small size of spines makes detailed measurement of their morphology in living cells best suited to super-resolution imaging techniques. The distribution of molecular positions mapped via live-cell Photoactivated Localization Microscopy (PALM) is a powerful approach, but molecular motion complicates this analysis and can degrade overall resolution of the morphological reconstruction. Nevertheless, the motion is of additional interest because tracking single molecules provides diffusion coefficients, bound fraction, and other key functional parameters. We used Monte Carlo simulations to examine features of single-molecule tracking of practical utility for the simultaneous determination of cell morphology. We find that the accuracy of determining both distance and angle of motion depend heavily on the precision with which molecules are localized. Strikingly, diffusion within a bounded region resulted in an inward bias of localizations away from the edges, inaccurately reflecting the region structure. This inward bias additionally resulted in a counterintuitive reduction of measured diffusion coefficient for fast-moving molecules; this effect was accentuated by the long camera exposures typically used in single-molecule tracking. Thus, accurate determination of cell morphology from rapidly moving molecules requires the use of short integration times within each image to minimize artifacts caused by motion during image acquisition. Sequential imaging of neuronal processes using excitation pulses of either 2 ms or 10 ms within imaging frames confirmed this: processes appeared erroneously thinner when imaged using the longer excitation pulse. Using this pulsed excitation approach, we show that PALM can be used to image spine and spine neck morphology in living neurons. These results clarify a number of issues involved in interpretation of single-molecule data in living cells and provide a method to minimize artifacts in single-molecule experiments.     

Comparison of Cartesian and Non-Cartesian Real-Time MRI Sequences at 1.5T to Assess Velar Motion and Velopharyngeal Closure during Speech

Dynamic imaging of the vocal tract using real-time MRI has been an active and growing area of research, having demonstrated great potential to become routinely performed in the clinical evaluation of speech and swallowing disorders. Although many technical advances have been made in regards to acquisition and reconstruction methodologies, there is still no consensus in best practice protocols. This study aims to compare Cartesian and non-Cartesian real-time MRI sequences, regarding image quality and temporal resolution trade-off, for dynamic speech imaging. Five subjects were imaged at 1.5T, while performing normal phonation, in order to assess velar motion and velopharyngeal closure. Data was acquired using both Cartesian and non-Cartesian (spiral and radial) real-time sequences at five different spatial-temporal resolution sets, between 10 fps (1.7×1.7×10 mm³) and 25 fps (1.5×1.5×10 mm³). Only standard scanning resources provided by the MRI scanner manufacturer were used to ensure easy applicability to clinical evaluation and reproducibility. Data sets were evaluated by comparing measurements of the velar structure, dynamic contrast-to-noise ratio and image quality visual scoring. Results showed that for all proposed sequences, FLASH spiral acquisitions provided higher contrast-to-noise ratio, up to a 170.34% increase at 20 fps, than equivalent bSSFP Cartesian acquisitions for the same spatial-temporal resolution. At higher frame rates (22 and 25 fps), spiral protocols were optimal and provided higher CNR and visual scoring than equivalent radial protocols. Comparison of dynamic imaging at 10 and 22 fps for radial and spiral acquisitions revealed no significant difference in CNR performance, thus indicating that temporal resolution can be doubled without compromising spatial resolution (1.9×1.9 mm²) or CNR. In summary, this study suggests that the use of FLASH spiral protocols should be preferred over bSSFP Cartesian for the dynamic imaging of velopharyngeal closure, as it allows for an improvement in CNR and overall image quality without compromising spatial-temporal resolution.     

Interpolated Compressed Sensing for 2D Multiple Slice Fast MR Imaging

Sparse MRI has been introduced to reduce the acquisition time and raw data size by undersampling the k-space data. However, the image quality, particularly the contrast to noise ratio (CNR), decreases with the undersampling rate. In this work, we proposed an interpolated Compressed Sensing (iCS) method to further enhance the imaging speed or reduce data size without significant sacrifice of image quality and CNR for multi-slice two-dimensional sparse MR imaging in humans. This method utilizes the k-space data of the neighboring slice in the multi-slice acquisition. The missing k-space data of a highly undersampled slice are estimated by using the raw data of its neighboring slice multiplied by a weighting function generated from low resolution full k-space reference images. In-vivo MR imaging in human feet has been used to investigate the feasibility and the performance of the proposed iCS method. The results show that by using the proposed iCS reconstruction method, the average image error can be reduced and the average CNR can be improved, compared with the conventional sparse MRI reconstruction at the same undersampling rate.     

Ultrasound-driven cardiac MRI

PurposeOne of the challenges of cardiac MR imaging is the compensation of respiratory motion, which causes the heart and the surrounding tissues to move. Commonly-used methods to overcome this effect, breath-holding and MR navigation, present shortcomings in terms of available acquisition time or need to periodically interrupt the acquisition, respectively. In this work, an implementation of respiratory motion compensation that obtains information from abdominal ultrasound and continuously adapts the imaged slice position in real time is presented.MethodsA custom workflow was developed, comprising an MR-compatible ultrasound acquisition system, a feature-motion-tracking system with polynomial predictive capability, and a custom MR sequence that continuously adapts the position of the acquired slice according to the tracked position. The system was evaluated on a moving phantom by comparing sharpness and image blurring between static and moving conditions, and in vivo by tracking the motion of the blood vessels of the liver to estimate the cardiac motion. Cine images of the heart were acquired during free breathing.ResultsIn vitro, the predictive motion correction yielded significantly better results than non-predictive or non-corrected acquisitions (p ≪ 0.01). In vivo, the predictive correction resulted in an image quality very similar to the breath-hold acquisition, whereas the uncorrected images show noticeable blurring artifacts.ConclusionIn this work, the possibility of using ultrasound navigation with tracking for the real-time adaptation of MR imaging slices was demonstrated. The implemented technique enabled efficient imaging of the heart with resolutions that would not be feasible in a single breath-hold.     

Evaluation of residual abdominal tumour motion in carbon ion gated treatments through respiratory motion modelling

At the Italian National Centre for Oncologic Hadrontherapy (CNAO) patients with upper-abdominal tumours are being treated with carbon ion therapy, adopting the respiratory gating technique in combination with layered rescanning and abdominal compression to mitigate organ motion. Since online imaging of the irradiated volume is not feasible, this study proposes a modelling approach for the estimation of residual motion of the target within the gating window. The model extracts a priori respiratory motion information from the planning 4DCT using deformable image registration (DIR), then combines such information with the external surrogate signal recorded during dose delivery. This provides estimation of a CT volume corresponding to any given respiratory phase measured during treatment. The method was applied for the retrospective estimation of tumour residual motion during irradiation, considering 16 patients treated at CNAO with the respiratory gating protocol. The estimated tumour displacement, calculated with respect to the reference end-exhale position, was always limited (average displacement is 0.32 ± 0.65 mm over all patients) and below the maximum motion defined in the treatment plan. This supports the hypothesis of target position reproducibility, which is the crucial assumption in the gating approach. We also demonstrated the use of the model as a simulation tool to establish a patient-specific relationship between residual motion and the width of the gating window. In conclusion, the implemented method yields an estimation of the repeatability of the internal anatomy configuration during gated treatments, which can be used for further studies concerning the dosimetric impact of the estimated residual organ motion.     

Sequential Superresolution Imaging of Multiple Targets Using a Single Fluorophore

Fluorescence superresolution (SR) microscopy, or fluorescence nanoscopy, provides nanometer scale detail of cellular structures and allows for imaging of biological processes at the molecular level. Specific SR imaging methods, such as localization-based imaging, rely on stochastic transitions between on (fluorescent) and off (dark) states of fluorophores. Imaging multiple cellular structures using multi-color imaging is complicated and limited by the differing properties of various organic dyes including their fluorescent state duty cycle, photons per switching event, number of fluorescent cycles before irreversible photobleaching, and overall sensitivity to buffer conditions. In addition, multiple color imaging requires consideration of multiple optical paths or chromatic aberration that can lead to differential aberrations that are important at the nanometer scale. Here, we report a method for sequential labeling and imaging that allows for SR imaging of multiple targets using a single fluorophore with negligible cross-talk between images. Using brightfield image correlation to register and overlay multiple image acquisitions with ~10 nm overlay precision in the x-y imaging plane, we have exploited the optimal properties of AlexaFluor647 for dSTORM to image four distinct cellular proteins. We also visualize the changes in co-localization of the epidermal growth factor (EGF) receptor and clathrin upon EGF addition that are consistent with clathrin-mediated endocytosis. These results are the first to demonstrate sequential SR (s-SR) imaging using direct stochastic reconstruction microscopy (dSTORM), and this method for sequential imaging can be applied to any superresolution technique.     

Artificial intelligence in image reconstruction: The change is here

Innovations in CT have been impressive among imaging and medical technologies in both the hardware and software domain. The range and speed of CT scanning improved from the introduction of multidetector-row CT scanners with wide-array detectors and faster gantry rotation speeds. To tackle concerns over rising radiation doses from its increasing use and to improve image quality, CT reconstruction techniques evolved from filtered back projection to commercial release of iterative reconstruction techniques, and recently, of deep learning (DL)-based image reconstruction. These newer reconstruction techniques enable improved or retained image quality versus filtered back projection at lower radiation doses. DL can aid in image reconstruction with training data without total reliance on the physical model of the imaging process, unique artifacts of PCD-CT due to charge sharing, K-escape, fluorescence x-ray emission, and pulse pileups can be handled in the data-driven fashion. With sufficiently reconstructed images, a well-designed network can be trained to upgrade image quality over a practical/clinical threshold or define new/killer applications. Besides, the much smaller detector pixel for PCD-CT can lead to huge computational costs with traditional model-based iterative reconstruction methods whereas deep networks can be much faster with training and validation. In this review, we present techniques, applications, uses, and limitations of deep learning-based image reconstruction methods in CT.     

Visualisation of Respiratory Tumour Motion and Co-Moving Isodose Lines in the Context of Respiratory Gating,IMRT and Flattening-Filter-Free Beams

Respiratory motion during percutaneous radiotherapy can be considered based on respiration-correlated computed tomography (4DCT). However, most treatment planning systems perform the dose calculation based on a single primary CT data set, even though cine mode displays may allow for a visualisation of the complete breathing cycle. This might create the mistaken impression that the dose distribution were independent of tumour motion. We present a movie visualisation technique with the aim to direct attention to the fact that the dose distribution migrates to some degree with the tumour and discuss consequences for gated treatment, IMRT plans and flattening-filter-free beams. This is a feasibility test for a visualisation of tumour and isodose motion. Ten respiratory phases are distinguished on the CT, and the dose distribution from a stationary IMRT plan is calculated on each phase, to be integrated into a movie of tumour and dose motion during breathing. For one example patient out of the sample of five lesions, the plan is compared with a gated treatment plan with respect to tumour coverage and lung sparing. The interplay-effect for small segments in the IMRT plan is estimated. While the high dose rate, together with the cone-shaped beam profile, makes the use of flattening-filter-free beams more problematic for conformal and IMRT treatment, it can be the option of choice if gated treatment is preferred. The different effects of respiratory motion, dose build-up and beam properties (segments and flatness) for gated vs. un-gated treatment can best be considered if planning is performed on the full 4DCT data set, which may be an incentive for future developments of treatment planning systems.     

Usefulness of a new online patient-specific quality assurance system for respiratory-gated radiotherapy

PurposeThe accuracy of gated irradiation may decrease when treatment is performed with short “beam-on” times. Also, the dose is subject to variation between treatment sessions if the respiratory rate is irregular. We therefore evaluated the impact of the differences between gated and non-gated treatment on doses using a new online quality assurance (QA) system for respiratory-gated radiotherapy.MethodsWe generated dose estimation models to associate dose and pulse information using a 0.6 cc Farmer chamber and our QA system. During gated irradiation with each of seven regular and irregular respiratory patterns, with the Farmer chamber readings as references, we evaluated our QA system’s accuracy. We then used the QA system to assess the impact of respiratory patterns on dose distribution for three lung and three liver radiotherapy plans. Gated and non-gated plans were generated and compared.ResultsThere was agreement within 1.7% between the ionization chamber and our system for several regular and irregular motion patterns. For dose distributions with measured errors, there were larger differences between gated and non-gated treatment for high-dose regions within the planned treatment volume (PTV). Compared with a non-gated plan, PTV D_95% for a gated plan decreased by −1.5% to −2.6%. Doses to organs at risk were similar with both plans.ConclusionsOur simple system estimated the radiation dose to the patient using only pulse information from the linac, even during irregular respiration. The quality of gated irradiation for each patient can be verified fraction by fraction.     

An image analysis toolbox for high-throughput C. elegans assays

We present a toolbox for high-throughput screening of image-based Caenorhabditis elegans phenotypes. The image analysis algorithms measure morphological phenotypes in individual worms and are effective for a variety of assays and imaging systems. This WormToolbox is available through the open-source CellProfiler project and enables objective scoring of whole-worm high-throughput image-based assays of C. elegans for the study of diverse biological pathways that are relevant to human disease.     

Comparative assessment of bone pose estimation using Point Cluster Technique and OpenSim

Estimating the position of the bones from optical motion capture data is a challenge associated with human movement analysis. Bone pose estimation techniques such as the Point Cluster Technique (PCT) and simulations of movement through software packages such as OpenSim are used to minimize soft tissue artifact and estimate skeletal position; however, using different methods for analysis may produce differing kinematic results which could lead to differences in clinical interpretation such as a misclassification of normal or pathological gait. This study evaluated the differences present in knee joint kinematics as a result of calculating joint angles using various techniques. We calculated knee joint kinematics from experimental gait data using the standard PCT, the least squares approach in OpenSim applied to experimental marker data, and the least squares approach in OpenSim applied to the results of the PCT algorithm. Maximum and resultant RMS differences in knee angles were calculated between all techniques. We observed differences in flexion/extension, varus/valgus, and internal/external rotation angles between all approaches. The largest differences were between the PCT results and all results calculated using OpenSim. The RMS differences averaged nearly 5° for flexion/extension angles with maximum differences exceeding 15°. Average RMS differences were relatively small (< 1.08°) between results calculated within OpenSim, suggesting that the choice of marker weighting is not critical to the results of the least squares inverse kinematics calculations. The largest difference between techniques appeared to be a constant offset between the PCT and all OpenSim results, which may be due to differences in the definition of anatomical reference frames, scaling of musculoskeletal models, and/or placement of virtual markers within OpenSim. Different methods for data analysis can produce largely different kinematic results, which could lead to the misclassification of normal or pathological gait. Improved techniques to allow non-uniform scaling of generic models to more accurately reflect subject-specific bone geometries and anatomical reference frames may reduce differences between bone pose estimation techniques and allow for comparison across gait analysis platforms.