Computed Tomography Image Sequences Research Articles

Investigating the Influence of Inlet Relative Humidity on Polymer Electrolyte Membrane Fuel Cell Performance by Visualizing 4-D Water Distributions in Gas Diffusion Layers

The catastrophic effects of atmospheric greenhouse gases and the depletion of non-renewable resources has led to the urgency to develop clean, sustainable energy technologies to meet increasing energy demands. However, the intermittent nature of current renewable energy technologies warrants the acquisition of on-demand renewable energy through either energy production or storage methods. Polymer electrolyte membrane fuel cells (PEMFCs) are promising candidates for this task, as they utilize the most abundant element on Earth, hydrogen, to produce high amounts of power under rapid changes in load with little to no greenhouse gas emissions (1). Therefore, PEMFCs have great potential to help offset the negative impact of atmospheric pollution due to excessive carbon emissions. However, liquid water management issues associated with high power output of the fuel cell typically leads to reduced performance and durability of the fuel cell, and thereby hinders global implementation of these devices (2). To minimize these losses and improve GDL material designs, an understanding of the relationship between product liquid water distributions in the cathode GDL and transport properties of PEMFCs under varying operating conditions is highly valuable. Previous works have characterized the effect of operating temperature on liquid water pathways and distributions in GDLs by visualizing operando PEMFCs with 3D imaging techniques such as X-ray computed tomography (CT) (3). The high-speed, high-resolution capabilities of these imaging techniques enable the visualization of dynamic pore-scale activity to elucidate transport mechanisms in the GDL.In this work, the effect of inlet relative humidity on the formation and distribution of liquid water pathways in cathode GDLs is investigated by imaging a PEMFC operando with synchrotron X-ray CT at high spatial resolution, enabling the resolution of water in the individual pores of the GDL. The contribution of a microporous layer (MPL) is also explored by imaging a cell with an MPL and without. Imaging is conducted on a specialized cell designed to facilitate continuous rotation about the CT stage, enabling fast acquisition of consecutive scans to achieve high temporal resolution useful for visualizing the dynamic development of preferential water pathways. Additionally, electrochemical impedance spectroscopy was performed to quantify mass transport losses. The sequence of CT images was utilized to capture the dynamics of liquid water development as well as stabilized water distributions. Visualizing the reconstructed images shows that as current increases, water production increases, and the development of water pathways to breakthrough at the flow field interface is observed. Additionally, results show that an increase in relative humidity led to a significant increase in cathode GDL water saturation.The contribution of this study to understanding transport mechanisms in GDL materials is significant to the characterization and optimal design of materials for improved PEMFC performance. Ultimately, the goal of this work is to accelerate the worldwide adoption of PEMFCs as a sustainable, reliable solution to replace conventional carbon-emitting energy sources.1. Alaswad et al., J. Hydrog. Energy, 41, (2016)2. Nagai et al., J. Power Sources, 435, (2019)3. D. Shum et al., Electrochem. Acta, 256, (2017)

An open-source application to identify the three-dimensional locations of electrodes implanted into the rat brain from computed tomography images

Electrophysiological recordings using metal electrodes implanted into the brains have been widely utilized to evaluate neuronal circuit dynamics related to behavior and external stimuli. The most common method for identifying implanted electrode tracks in the brain tissue has been histological examination following postmortem slicing and staining of the brain tissue, which consumes time and resources and occasionally fails to identify the tracks because the brain preparations have been damaged during processing. Recent studies have proposed the use of a promising alternative method, consisting of computed tomography (CT) scanning that can directly reconstruct the three-dimensional arrangements of electrodes in the brains of living animals. In this study, we developed an open-source Python-based application that estimates the location of an implanted electrode from CT image sequences in a rat. After the user manually sets reference coordinates and an area from a sequence of CT images, this application automatically overlays an estimated location of an electrode tip on a histological template image; the estimates are highly accurate, with less than 135 µm of error, irrespective of the depth of the brain region. The estimation of an electrode location can be completed within a few minutes. Our simple and user-friendly application extends beyond currently available CT-based electrode localization methods and opens up the possibility of applying this technique to various electrophysiological recording paradigms.

Deep 3D reconstruction of synchrotron X-ray computed tomography for intact lungs

Synchrotron X-rays can be used to obtain highly detailed images of parts of the lung. However, micro-motion artifacts induced by such as cardiac motion impede quantitative visualization of the alveoli in the lungs. This paper proposes a method that applies a neural network for synchrotron X-ray Computed Tomography (CT) data to reconstruct the high-quality 3D structure of alveoli in intact mouse lungs at expiration, without needing ground-truth data. Our approach reconstructs the spatial sequence of CT images by using a deep-image prior with interpolated input latent variables, and in this way significantly enhances the images of alveolar structure compared with the prior art. The approach successfully visualizes 3D alveolar units of intact mouse lungs at expiration and enables us to measure the diameter of the alveoli. We believe that our approach helps to accurately visualize other living organs hampered by micro-motion.

Shortcomings of Posterior Malleolus Fractures Classifications: A Independent Inter and Intra-Rater Agreement Study Was Proposed

Category: Ankle Introduction/Purpose: Posterior malleolus fractures (PMFs) occur up to 50% of ankle fractures. Several classification systems exist for their characterization, specially under computed tomography (CT). However, those classifications do not report or report incompletely the level of agreement and reliability. Also, there are no studies that have validated those classifications further. The aim of this study was to perform an independent agreement study for comparing the level of agreement of Haraguchi, Bartonícek/Rammelt, and Mason's classification; among physicians with different levels of orthopaedic training in the management of PMFs. Methods: This study was designed according to the 'Guidelines for Reporting Reliability and Agreement Studies' ('GRASS'). Ninety-four CT scans of patients with ankle fractures that had PMF were included. Exclusion criteria were previous history of fracture, surgery, tumors, and infections of the ankle; and the presence of pilon fracture. PMF were assessed by six evaluators (3 foot and ankle surgeons and 3 orthopaedic surgery residents) according to Haraguchi, Bartonícek/Rammelt; and Mason 's classifications. All images were reassessed by the same evaluators in a random sequence three weeks later. The new sequence of CT images was done by one of the authors, who did not participate in the evaluation of the cases. The Kappa (k) oefficient was used to determine the intra and interobserver agreement. Statistical significance was established using p < .05, with a 95% confidence interval (CI). Results: We performed a total of 564 observations for each classification. The overall interobserver agreement was moderate for all classifications, with a (k) value of 0.52 (95% CI: 0.43-0.61) for Haraguchi, 0.53 (95% CI: 0.46-0.61) for Bartonícek/Rammelt; and 0.54 (95% CI: 0.47-0.62) for Mason's classification, without significant differences between them. Meanwhile, the overall interobserver agreement was substantial for all classifications, with a (k) value of 0.70 (95% CI: 0.64-0.75) for Haraguchi, 0.73 (95% CI: 0.69-0.78) for Bartonícek/Rammelt; and 0.73 (95% CI: 0.69-0.78) for Mason's classification, without significant differences between them. No differences were found in the inter and intraobserver agreement among orthopaedic surgeons and residents. Conclusion: This is the first agreement study that compares the most utilized classifications for PMFs, according to 'GRRAS'. Our results showed substantial intraobserver agreement, but only moderate interobserver agreement for all classifications; limiting adequate communication in clinical and research field. These findings did not change regardless of the level of experience of the evaluators. Due to moderate level of agreement and that no classification is superior to the others, we estimate the need to generate a new classification for PMFs, that gives a higher level of agreement, as well as, orientation to the surgeon during the decision-making process.

Progressive Distributed and Parallel Similarity Retrieval of Large CT Image Sequences in Mobile Telemedicine Networks

Computed tomography image (CTI) sequence is essentially a time-series data that typically consists of a large amount of nearby and similar CTIs. Due to the high communication and computational costs, it is difficult to perform a progressive distributed similarity retrieval of the large CTI sequence (CTIS)s, particularly in resource-constraint mobile telemedicine network (MTN)s. In this paper, we present a Dprs method—progressive distributed and parallel similarity retrieval scheme for the CTISs in the MTN. To the best of our knowledge, there is little research on the Dprs processing, especially in the MTN. Four supporting techniques (i.e., (1) PCTI-based similarity measurement, (2) lightweight privacy-preserving strategy, (3) SSL-based data distribution scheme, and (4) the UDI framework) are developed. The experimental evaluation indicates that our proposed Dprs method is more progressive than the state of the art, with a significant reduction in response time.

Reliable detection of lymph nodes in whole pelvic for radiotherapy

Accurate detection of the lymph nodes in patients with cervical cancer enables more targeted treatment. For drainage or radiotherapy of postoperative adjuvant, clinicians will manually label the lymph nodes in Computed tomography (CT) images. However, due to the complex environment and wide area of pelvic, reliable detection of lymph nodes is challenging because of the changeable shapes, sizes, distributions. Furthermore, the confusion of blood vessels and tissues in CT images also causes difficulties when labeling lymph nodes. In this study, a reliable convolutional neural network (CNN) based detection approach is proposed to distinguish lymph nodes in the middle abdomen and pelvic cavity between CT image sequences. Combining both local and global contextual information in CT image sequences can avoiding complex three-dimensional (3D) computation. Different scale transformation and structural similarity-based processing method ensure the accuracy to distinguish lymph nodes and blood vessels. Experimental results compared with those given by radiologists and other CNN based methods proved that our proposed method could locate lymph nodes with 98.29% accuracy and 94.64% recall rate when detecting 22,846 clinical abdominal CT images, which proved its potential application value for radiotherapy of cervical cancer.

Multi-View Auxiliary Diagnosis Algorithm for Lung Nodules

Lung is an important organ of human body. More and more people are suffering from lung diseases due to air pollution. These diseases are usually highly infectious. Such as lung tuberculosis, novel coronavirus COVID-19, etc. Lung nodule is a kind of high-density globular lesion in the lung. Physicians need to spend a lot of time and energy to observe the computed tomography image sequences to make a diagnosis, which is inefficient. For this reason, the use of computer-assisted diagnosis of lung nodules has become the current main trend. In the process of computer-aided diagnosis, how to reduce the false positive rate while ensuring a low missed detection rate is a difficulty and focus of current research. To solve this problem, we propose a three-dimensional optimization model to achieve the extraction of suspected regions, improve the traditional deep belief network, and to modify the dispersion matrix between classes. We construct a multi-view model, fuse local three-dimensional information into two-dimensional images, and thereby to reduce the complexity of the algorithm. And alleviate the problem of unbalanced training caused by only a small number of positive samples. Experiments show that the false positive rate of the algorithm proposed in this paper is as low as 12%, which is in line with clinical application standards.

Liver segmentation in CT imaging with enhanced mask region-based convolutional neural networks.

BackgroundLiver segmentation in computed tomography (CT) imaging has been widely investigated as a crucial step for analyzing liver characteristics and diagnosing liver diseases. However, obtaining satisfactory liver segmentation performance is highly challenging because of the poor contrast between the liver and its surrounding organs and tissues, the high levels of CT image noise, and the wide variability in liver shapes among patients.MethodsTo overcome these challenges, we propose a novel method for liver segmentation in CT image sequences. This method uses an enhanced mask region-based convolutional neural network (Mask R-CNN) with graph-cut segmentation. Specifically, the k-nearest neighbor (k-NN) algorithm is employed to cluster the target liver pixels in order to get an appropriate aspect ratio. Then, anchors are adapted to the liver size using the ratio information. Thus, high-accuracy liver localization can be achieved using the anchors and rotation-invariant object recognition. Next, a fully convolutional network (FCN) is used to segment the foreground objects, and local fine-grained liver detection is realized by pixel prediction. Finally, a whole liver mask is obtained by Mask R-CNN proposed in this paper.ResultsWe proposed a Mask R-CNN algorithm which achieved superior performance in comparison with the conventional Mask R-CNN algorithms in term of the dice similarity coefficient (DSC), and the Medical Image Computing and Computer-Assisted Intervention (MICCAI) metrics.ConclusionsOur experimental results demonstrate that the improved Mask R-CNN architecture has good performance, accuracy, and robustness for liver segmentation in CT image sequences.

Automated Classification of Papillary Renal Cell Carcinoma and Chromophobe Renal Cell Carcinoma Based on a Small Computed Tomography Imaging Dataset Using Deep Learning.

ObjectivesThis study was conducted in order to design and develop a framework utilizing deep learning (DL) to differentiate papillary renal cell carcinoma (PRCC) from chromophobe renal cell carcinoma (ChRCC) using convolutional neural networks (CNNs) on a small set of computed tomography (CT) images and provide a feasible method that can be applied to light devices.MethodsTraining and validation datasets were established based on radiological, clinical, and pathological data exported from the radiology, urology, and pathology departments. As the gold standard, reports were reviewed to determine the pathological subtype. Six CNN-based models were trained and validated to differentiate the two subtypes. A special test dataset generated with six new cases and four cases from The Cancer Imaging Archive (TCIA) was applied to validate the efficiency of the best model and of the manual processing by abdominal radiologists. Objective evaluation indexes [accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curve, and area under the curve (AUC)] were calculated to assess model performance.ResultsThe CT image sequences of 70 patients were segmented and validated by two experienced abdominal radiologists. The best model achieved 96.8640% accuracy (99.3794% sensitivity and 94.0271% specificity) in the validation set and 100% (case accuracy) and 93.3333% (image accuracy) in the test set. The manual classification achieved 85% accuracy (100% sensitivity and 70% specificity) in the test set.ConclusionsThis framework demonstrates that DL models could help reliably predict the subtypes of PRCC and ChRCC.

Mask R-CNN-based feature extraction and three-dimensional recognition of rice panicle CT images.

The rice panicle seed setting rate is extremely important for calculating rice yield and performing genetic analysis. Unlike machine vision, X‐ray computed tomography (CT) imaging is a nondestructive technique that provides direct information on the internal and external structure of rice panicles. However, occlusion and adhesion of panicles and grains in a CT image sequence make these objects difficult to identify, which in turn hinders accurate determination of the seed setting rate of rice panicles. Therefore, this paper proposes a method based on a mask region convolutional neural network (Mask R‐CNN) for feature extraction and three‐dimensional (3‐D) recognition of CT images of rice panicles. X‐ray CT feature characterization was combined with the Mask R‐CNN algorithm to perform feature extraction and classification of a panicle and grains in each layer of the CT sequence. The Euclidean distance between adjacent layers was minimized to extract the features of a 3‐D panicle and grains. The results were used to calculate the rice panicle seed setting rate. The proposed method was experimentally verified using eight sets of different rice panicles. The results showed that the proposed method can efficiently identify and count plump grains and blighted grains to achieve an accuracy above 99% for the seed setting rate.

Study on strategy of CT image sequence segmentation for liver and tumor based on U-Net and Bi-ConvLSTM

Accurate segmentation of the liver and tumors in computed tomography (CT) images is critical for intelligent computer-aided diagnosis (CAD). The commonly used segmentation methods based on fully convolutional networks (FCN) only take a single image into consideration but do not make good use of sequence information. In this paper, two more feasible sequence segmentation strategies than 3D U-net which can utilize inter-slice and intra-slice features simultaneously at the lower hardware and time cost are studied to improve the segmentation result. U-net serves as the backbone model of segmentation and Bi-directional convolutional long short-term memory (Bi-ConvLSTM) is chosen to extract and fuse the inter-slice feature. Strategy A corrects the pre-segmented results of U-net in the fusion of sequence information as a post-processing, where Mod-1, Mod-2 and Mod-3 models are built to compare the effects of width, depth, and residual structure on the modified model of sequence segmentation. Strategy B directly integrates the fusion of sequence information into the feature extraction of U-net, and then an end-to-end model called W-net is built based on it. The experiment results show that both strategies improve the liver and tumor segmentation performance in various aspects. The results based on strategy A are closer to the ground truth with less misdiagnose region: Mod-1 achieves better accuracy on liver contour segmentation because of the largest model width; Mod-2 can obtain more accurate tumor contour since the greatest depth of feature extraction process; and Mod-3 is at the average segmentation performance. Therefore, strategy A is recommended in the application of surgery planning of tumors. Strategy B achieves better space coincidence degree and less training time cost, which is more suitable for the early screening of liver cancer.

Automatic tooth roots segmentation of cone beam computed tomography image sequences using U-net and RNN.

Automatic segmentation of individual tooth root is a key technology for the reconstruction of the three-dimensional dental model from Cone Beam Computed Tomography (CBCT) images, which is of great significance for the orthodontic, implant and other dental diagnosis and treatment planning. Currently, tooth root segmentation is mainly done manually because of the similar gray of the tooth root and the alveolar bone from CBCT images. This study aims to explore the automatic tooth root segmentation algorithm of CBCT axial image sequence based on deep learning. We proposed a new automatic tooth root segmentation method based on the deep learning U-net with AGs. Since CBCT sequence has a strong correlation between adjacent slices, a Recurrent neural network (RNN) was applied to extract the intra-slice and inter-slice contexts. To develop and test this new method for automatic segmentation of tooth roots using CBCT images, 24 sets of CBCT sequences containing 1160 images and 5 sets of CBCT sequences containing 361 images were used to train and test the network, respectively. Applying to the testing dataset, the segmentation accuracy measured by the intersection over union (IOU), dice similarity coefficient (DICE), average precision rate (APR), average recall rate (ARR), and average symmetrical surface distance (ASSD) are 0.914, 0.955, 95.8% , 95.3% , 0.145 mm, respectively. The study demonstrates that the new method combining attention U-net with RNN yields the promising results of automatic tooth roots segmentation, which has potential to help improve the segmentation efficiency and accuracy in future clinical practice.

Fracture Segmentation Method Based on Contour Evolution and Gradient Direction Consistency in Sequence of Coal Rock CT Images

The coal rock exhibits obvious heterogeneity and anisotropy after a long-term geological evolution in nature. Therefore, the coal rock CT (computed tomography) image shows uneven grey scale, low contrast, and the fractures with weak boundaries. The accurate segmentation of the coal rock fracture networks is challenging. In this paper, a segmentation method of fractures based on contour evolution and gradient direction consistency is proposed to accurately segment the fracture networks in the sequence of coal rock CT images. According to the contour variation rule of the fractures in the discrete 3D (three dimensional) space formed by the sequence of CT images, the fracture contour evolution model (FCEM) is constructed and the preliminary segmentation results of fractures are obtained from FCEM. A 3D adaptive median filtering (3DAMF) and a 3D bilateral filtering (3DBF) are proposed. The high density miscellaneous point noises in the coal rock CT images are filtered by the 3DAMF. And the boundaries of fractures are enhanced by 3DBF. According to the similarity of the preliminary segmentation results of fractures and the real contours of fractures, the preliminary segmentation results of fractures are optimized based on the gradient direction consistency model (GDCM) proposed in this paper to obtain the accurate boundaries of fractures. The fracture segmentation method proposed in this paper can obtain accurate boundaries of fractures with weak boundaries, and the experimental results show that the segmentation efficiency for sequence is high and adaptability is strong.

A Novel Technique to Assess Distal Radioulnar Joint Stability Using Increasing Torque.

Background Previous studies on computed tomography (CT) in patients with a suspected triangular fibrocartilage complex (TFCC) injury have not been successful in assessing distal radioulnar joint (DRUJ) laxity. The aim of this study was to develop a novel servomotor-driven device for the assessment of DRUJ by applying increasing torque to the DRUJ in pronation and supination. Methods A custom-built device was designed to function during four-dimensional (4D) CT of the wrist. A torque meter, positioned between the incoming hand holder, and a direct current (DC) servomotor were used for angular positioning and for applying rotational force to the patient's arm. A total of 110 healthy participants were recruited to gather reference values for the range of motion (ROM), maximum torque in neutral and supinated/pronated position, and the ability to withstand an increasing, device-generated torque in these positions. The device was also used during 4D DRUJ CT in five patients with suspected TFCC injuries. Results A gender- and age-relevant reference chart for ROM and torque was created. Men showed a tendency (ns) toward having a larger ROM and increasing strength with increasing age, whereas women showed the opposite. Also, the dominant hand showed a tendency toward having a larger ROM and being stronger than the nondominant hand (ns). A smaller cohort of patients ( n = 5) with suspected TFCC injuries showed a significantly decreased ability to withstand increasing torque in both supination (2.1 ± 0.3 vs. 3.1 ± 0.2 s; p < 0.005) and pronation (2.3 ± 0.5 vs. 3.1 ± 0.4 s; p < 0.0005) and also showed a clear laxity on real-time 4D CT image sequences. Decreased strength at all positions was also found (average 74% decrease compared to noninjured side). Conclusion Reference values for torque strength and ability to withstand increasing torque can be used clinically in the assessment of patients with symptoms that could represent ligamentous injuries to the TFCC. The ability to use the device during CT enables radiographic evaluation of instability during increasing torque. Level of Evidence This is a Level II study.

Quantifying fibre reorientation during axial compression of a composite through time-lapse X-ray imaging and individual fibre tracking

The sudden compressive failure of unidirectional (UD) fibre reinforced composites at loads well below their tensile strengths is a cause of practical concern. In this respect and more generally, analytical and numerical models that describe composite behaviour have been hard to verify due to a lack of experimental observation, particularly in 3D. The aim of this paper is to combine fast in-situ X-ray computed tomography (CT) with advanced image analysis to capture the changes in fibre orientation in 3D during uninterrupted progressive loading in compression of a UD glass fibre reinforced polymer (GFRP). By analysing and establishing correspondence between a sequence of time-lapse X-ray CT images of the composite, we are able for the first time to follow each fibre and quantify the progressive deflection that takes place during axial compression in the steps leading up to fibre micro-buckling and kinking. Even at just 25% of the failure load, fibres have started to tilt in approximately the direction of the ultimate kink band. The rate of tilting increases as the composite approaches the collapse load. More generally, our approach can be applied to investigate the behaviour of a wide range of fibrous materials under changing loading conditions.

The segmentation of bones in pelvic CT images based on extraction of key frames

BackgroundBone segmentation is important in computed tomography (CT) imaging of the pelvis, which assists physicians in the early diagnosis of pelvic injury, in planning operations, and in evaluating the effects of surgical treatment. This study developed a new algorithm for the accurate, fast, and efficient segmentation of the pelvis.MethodsThe proposed method consists of two main parts: the extraction of key frames and the segmentation of pelvic CT images. Key frames were extracted based on pixel difference, mutual information and normalized correlation coefficient. In the pelvis segmentation phase, skeleton extraction from CT images and a marker-based watershed algorithm were combined to segment the pelvis. To meet the requirements of clinical application, physician’s judgment is needed. Therefore the proposed methodology is semi-automated.ResultsIn this paper, 5 sets of CT data were used to test the overlapping area, and 15 CT images were used to determine the average deviation distance. The average overlapping area of the 5 sets was greater than 94%, and the minimum average deviation distance was approximately 0.58 pixels. In addition, the key frame extraction efficiency and the running time of the proposed method were evaluated on 20 sets of CT data. For each set, approximately 13% of the images were selected as key frames, and the average processing time was approximately 2 min (the time for manual marking was not included).ConclusionsThe proposed method is able to achieve accurate, fast, and efficient segmentation of pelvic CT image sequences. Segmentation results not only provide an important reference for early diagnosis and decisions regarding surgical procedures, they also offer more accurate data for medical image registration, recognition and 3D reconstruction.

Combining X-ray microtomography and three-dimensional digital volume correlation to track microstructure evolution during sintering of copper powder

Optimising the manufacture, and ultimately the mechanical performance, of powder-processed components requires an understanding of how the state of the material evolves during processing and in particular during the final sintering stage. Synchrotron X-ray microtomography has been employed to follow in situ the evolution of particle microstructure during sintering of copper powder. In particular, three-dimensional digital volume correlation of a time-lapse computed tomography image sequence allows accurate quantification of the three-dimensional movements of the particles and thereby local strain for the first time. Strains are quantified both at a coarse scale across the whole powder assembly and at higher magnification for a smaller local region of interest. Unsurprisingly, the rate of shrinkage is observed to decrease with sintering time in accordance with changes in the overall density. Heterogeneities in straining within the body are observed at the several particle level, often associated with anisotropic shrinkage in part associated with non-uniform movement and densification of small aggregates of particles between which significant changes in shrinkage are observed. These differences in shrinkage could be responsible for cracking on further densification.

Inter-rater accuracy and precision of markerless quantification of 3d tibiofemoral displacement using dual fluoroscopy

Purpose: Early changes in osteoarthritic (OA) knee cartilage include tissue swelling and softening, which may predispose cartilage to mechanical damage during activities of daily living. Current clinical OA diagnosis relies on radiographic bone morphology change and pain, which can be insensitive to early change. Developing a marker that specifically targets early mechanical tissue changes may advance the current state of OA assessment. Potential candidates for such markers are in-vivo magnitude and rate of tibiofemoral (TF) soft tissue deformation during weight bearing (compression) and unloading (recovery). Respective joint “signatures” for healthy and OA groups may then allow for early OA identification. Dual Fluoroscopy (DF) is an emerging tool with the potential to provide the temporal and spatial resolutions necessary for such assessment. Tissue deformation mechanics may be derived using translation and rotation changes of the femur with respect to the tibia over time. Using a bead tracking technique, similar to radiostereometric analysis (RSA), translation changes of 0.05 mm (1/ 10th expected maximal cartilage deformation) have previously been accurately detected. It is however unknown if a non-invasive markerless DF approach is sufficiently accurate and precise for quantifying the small changes during weight bearing. The purpose of this study was to establish the Minimum Detectable Displacement (MDD) of 3D femoral translations with respect to the tibia in a custom built DF system using a markerless shape matching image registration approach. Methods: The DF system consisted of two X-ray tubes (Varian, USA) and generators (EMD, Canada), quad mode image intensifiers (Toshiba, Japan) and high speed digital video cameras (PCO, Germany). DF images were acquired with an entrance field diameter of 228 mm and 0.155 x 0.155 mm2 pixel resolution. Dry tibia and femur bones were fixed to a plexiglass frame, where the tibia remained stationary and the femur was translated using a digital micrometer (Mitutoyo, Japan). The frame was placed within the DF field-of-view and did not move during data capture. Ten displacement intervals (0.01-0.1 mm)were imagedwith 10 repetitions of each interval. DF calibration was performed using a calibration cube and modified direct linear transform. Images were distortion corrected using a perforated steel grid and XrayProject (Brown University, USA). 3D femur and tibia bone models were generated using segmentation (Amira, USA) of Computed Tomography image sequences (0.21 x 0.21 x 0.31 mm). 3D bone orientations were obtained by three raters using AutoScoper (Brown University, USA) and a markerless shape matching approach. Bone-specific models were positioned to align their visible features to those of the underlying 2D X-ray images. Relative femur and tibia displacements were obtained using the Euclidean distance between bone models. Independent samples t-tests were used to detect MDDs of the moving femur relative to the stationary tibia at each displacement interval. Accuracy and precision were computed as means and standard deviations of the absolute differences of the femur translations compared to the micrometer displacements. Inter-rater reliability values of the femur were computed using intra-class correlation coefficient (ICC). All statistical analyses were performed in SPSS (IBM, USA) and p 0.05. Results: Femur displacement measurements were significantly different from tibia (p<0.045) and had greater than 95% confidence in accuracy at micrometer displacements 0.:08mm (accuracy precision: 0.022 0.019) (Figure 1). Between raters femur displacement measurements were not significant for micrometer displacements 0.:05mm and inter-rater reliability across all displacements was 92%. Conclusions: These findings indicate that 3D bone translations within relevant in-vivo ranges can be observed accurately and precisely using a markerless DF shape matching approach. Femur displacements of 0.:08mm were accurately and reliably detected with respect to set micrometer displacements. The data further indicated good between rater agreement when quantifying femur displacements. Given the above system specifications, the markerless DF approach provides a means for non-invasive in-vivo quantification of TF tissue response to loading and unloading stimuli. Using this approach, further workwill be performed to establish the mechanical behaviour of TF soft tissues of healthy and ligament deficient and OA knees. While significant user input is required, compared to RSA type techniques, this non-invasive approach may be more readily applied in a future clinical setting.

Inter‐slice bidirectional registration‐based segmentation of the prostate gland in MR and CT image sequences

Accurate segmentation and volume estimation of the prostate gland in magnetic resonance (MR) and computed tomography (CT) images are necessary steps in diagnosis, treatment, and monitoring of prostate cancer. This paper presents an algorithm for the prostate gland volume estimation based on the semiautomated segmentation of individual slices in T2-weighted MR and CT image sequences. The proposed Inter-Slice Bidirectional Registration-based Segmentation (iBRS) algorithm relies on interslice image registration of volume data to segment the prostate gland without the use of an anatomical atlas. It requires the user to mark only three slices in a given volume dataset, i.e., the first, middle, and last slices. Next, the proposed algorithm uses a registration algorithm to autosegment the remaining slices. We conducted comprehensive experiments to measure the performance of the proposed algorithm using three registration methods (i.e., rigid, affine, and nonrigid techniques). The results with the proposed technique were compared with manual marking using prostate MR and CT images from 117 patients. Manual marking was performed by an expert user for all 117 patients. The median accuracies for individual slices measured using the Dice similarity coefficient (DSC) were 92% and 91% for MR and CT images, respectively. The iBRS algorithm was also evaluated regarding user variability, which confirmed that the algorithm was robust to interuser variability when marking the prostate gland. The proposed algorithm exploits the interslice data redundancy of the images in a volume dataset of MR and CT images and eliminates the need for an atlas, minimizing the computational cost while producing highly accurate results which are robust to interuser variability.

CT image sequence restoration based on sparse and low-rank decomposition.

Blurry organ boundaries and soft tissue structures present a major challenge in biomedical image restoration. In this paper, we propose a low-rank decomposition-based method for computed tomography (CT) image sequence restoration, where the CT image sequence is decomposed into a sparse component and a low-rank component. A new point spread function of Weiner filter is employed to efficiently remove blur in the sparse component; a wiener filtering with the Gaussian PSF is used to recover the average image of the low-rank component. And then we get the recovered CT image sequence by combining the recovery low-rank image with all recovery sparse image sequence. Our method achieves restoration results with higher contrast, sharper organ boundaries and richer soft tissue structure information, compared with existing CT image restoration methods. The robustness of our method was assessed with numerical experiments using three different low-rank models: Robust Principle Component Analysis (RPCA), Linearized Alternating Direction Method with Adaptive Penalty (LADMAP) and Go Decomposition (GoDec). Experimental results demonstrated that the RPCA model was the most suitable for the small noise CT images whereas the GoDec model was the best for the large noisy CT images.