A constrained linear regression optimization algorithm for diaphragm motion tracking with cone beam CT projections

PurposeWe presented a feasibility study to extract the diaphragm motion from the inferior contrast cone beam computed tomography (CBCT) projection images using a constrained linear regression optimization algorithm.MethodsThe shape of the diaphragm was fitted by a parabolic function which was initialized by five manually placed points on the diaphragm contour of a pre-selected projection. A constrained linear regression model by exploiting the spatial, algebraic, and temporal constraints of the diaphragm, approximated by a parabola, was employed to estimate the parameters. The algorithm was assessed by a fluoroscopic movie acquired at anterior-posterior (AP) fixed direction and kilovoltage CBCT projection image sets from four lung and two liver patients using the Varian 21iX Clinac. The automatic tracing by the proposed algorithm and manual tracking were compared in both space and frequency domains for the algorithm evaluations.ResultsThe error between the results estimated by the proposed algorithm and those by manual tracking for the AP fluoroscopic movie was 0.54 mm with standard deviation (SD) of 0.45 mm. For the detected projections the average error was 0.79 mm with SD of 0.64 mm for all six enrolled patients and the maximum deviation was 2.5 mm. The mean sub-millimeter accuracy outcome exhibits the feasibility of the proposed constrained linear regression approach to track the diaphragm motion on rotational fluoroscopic images.ConclusionThe new algorithm will provide a potential solution to rendering diaphragm motion and possibly aiding the tumor target tracking in radiation therapy of thoracic/abdominal cancer patients.     

Increases of calcium and magnesium and decrease of iron in human posterior longitudinal ligaments of the cervical spine with aging

To elucidate compositional changes of ligaments with aging, the authors investigated age-related changes of elements in the posterior longitudinal ligaments (PLLs) by inductively coupled plasma—atomic emission spectrometry. After the ordinary dissection, PLLs were resected from the subjects ranging in age from 65 to 95 yr. The PLLs of the cervical spine were resected between the fourth and fifth cervical vertebrae, the PLLs of the thoracic spine between the fifth and seventh thoracic vertebrae, and the PLLs of the lumbar spine between the second and third lumbar vertebrae. Calcium and magnesium increased progressively with aging in the PLLs of the cervical spine, but they did not increase with aging in the PLLs of the thoracic and lumbar spine. In contrast, iron decreased gradually with aging in the PLLs of the cervical spine. Regarding the relationships among elements, significant correlations were found among the contents of calcium, phosphorus, magnesium, and sodium in the PLLs of the cervical spine.     

Three-dimensional static modeling of the lumbar spine

This paper presents three-dimensional static modeling of the human lumbar spine to be used in the formation of anatomically-correct movement patterns for a fully cable-actuated robotic lumbar spine which can mimic in vivo human lumbar spine movements to provide better hands-on training for medical students. The mathematical model incorporates five lumbar vertebrae between the first lumbar vertebra and the sacrum, with dimensions of an average adult human spine. The vertebrae are connected to each other by elastic elements, torsional springs and a spherical joint located at the inferoposterior corner in the mid-sagittal plane of the vertebral body. Elastic elements represent the ligaments that surround the facet joints and the torsional springs represent the collective effect of intervertebral disc which plays a major role in balancing torsional load during upper body motion and the remaining ligaments that support the spinal column. The elastic elements and torsional springs are considered to be nonlinear. The nonlinear stiffness constants for six motion types were solved using a multiobjective optimization technique. The quantitative comparison between the angles of rotations predicted by the proposed model and in the experimental data confirmed that the model yields angles of rotation close to the experimental data. The main contribution is that the new model can be used for all motions while the experimental data was only obtained at discrete measurement points.     

A Novel Tracking Algorithm via Feature Points Matching

Visual target tracking is a primary task in many computer vision applications and has been widely studied in recent years. Among all the tracking methods, the mean shift algorithm has attracted extraordinary interest and been well developed in the past decade due to its excellent performance. However, it is still challenging for the color histogram based algorithms to deal with the complex target tracking. Therefore, the algorithms based on other distinguishing features are highly required. In this paper, we propose a novel target tracking algorithm based on mean shift theory, in which a new type of image feature is introduced and utilized to find the corresponding region between the neighbor frames. The target histogram is created by clustering the features obtained in the extraction strategy. Then, the mean shift process is adopted to calculate the target location iteratively. Experimental results demonstrate that the proposed algorithm can deal with the challenging tracking situations such as: partial occlusion, illumination change, scale variations, object rotation and complex background clutter. Meanwhile, it outperforms several state-of-the-art methods.     

Possibility Study of Scale Invariant Feature Transform (SIFT) Algorithm Application to Spine Magnetic Resonance Imaging

The purpose of this study is an application of scale invariant feature transform (SIFT) algorithm to stitch the cervical-thoracic-lumbar (C-T-L) spine magnetic resonance (MR) images to provide a view of the entire spine in a single image. All MR images were acquired with fast spin echo (FSE) pulse sequence using two MR scanners (1.5 T and 3.0 T). The stitching procedures for each part of spine MR image were performed and implemented on a graphic user interface (GUI) configuration. Moreover, the stitching process is performed in two categories; manual point-to-point (mPTP) selection that performed by user specified corresponding matching points, and automated point-to-point (aPTP) selection that performed by SIFT algorithm. The stitched images using SIFT algorithm showed fine registered results and quantitatively acquired values also indicated little errors compared with commercially mounted stitching algorithm in MRI systems. Our study presented a preliminary validation of the SIFT algorithm application to MRI spine images, and the results indicated that the proposed approach can be performed well for the improvement of diagnosis. We believe that our approach can be helpful for the clinical application and extension of other medical imaging modalities for image stitching.     

A comparison of denoising methods for X-ray fluoroscopic images

Fluoroscopic images exhibit severe signal-dependent quantum noise, due to the reduced X-ray dose involved in image formation, that is generally modelled as Poisson-distributed. However, image gray-level transformations, commonly applied by fluoroscopic device to enhance contrast, modify the noise statistics and the relationship between image noise variance and expected pixel intensity. Image denoising is essential to improve quality of fluoroscopic images and their clinical information content. Simple average filters are commonly employed in real-time processing, but they tend to blur edges and details. An extensive comparison of advanced denoising algorithms specifically designed for both signal-dependent noise (AAS, BM3Dc, HHM, TLS) and independent additive noise (AV, BM3D, K-SVD) was presented. Simulated test images degraded by various levels of Poisson quantum noise and real clinical fluoroscopic images were considered. Typical gray-level transformations (e.g. white compression) were also applied in order to evaluate their effect on the denoising algorithms. Performances of the algorithms were evaluated in terms of peak-signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), mean square error (MSE), structural similarity index (SSIM) and computational time. On average, the filters designed for signal-dependent noise provided better image restorations than those assuming additive white Gaussian noise (AWGN). Collaborative denoising strategy was found to be the most effective in denoising of both simulated and real data, also in the presence of image gray-level transformations. White compression, by inherently reducing the greater noise variance of brighter pixels, appeared to support denoising algorithms in performing more effectively.     

Finite element lumbar spine facet contact parameter predictions are affected by the cartilage thickness distribution and initial joint gap size

Current finite element modeling techniques utilize geometrically inaccurate cartilage distribution representations in the lumbar spine. We hypothesize that this shortcoming severely limits the predictive fidelity of these simulations. Specifically, it is unclear how these anatomically inaccurate cartilage representations alter range of motion and facet contact predictions. In the current study, cadaveric vertebrae were serially sectioned, and images were taken of each slice in order to identify the osteochondral interface and the articulating surface. A series of custom-written algorithms were utilized in order to quantify each facet joint's three-dimensional cartilage distribution using a previously developed methodology. These vertebrae-dependent thickness cartilage distributions were implemented on an L1 through L5 lumbar spine finite element model. Moments were applied in three principal planes of motion, and range of motion and facet contact predictions from the variable thickness and constant thickness distribution models were determined. Initial facet gap thickness dimensions were also parameterized. The data indicate that the mean and maximum cartilage thickness increased inferiorly from L1 to L5, with an overall mean thickness value of 0.57 mm. Cartilage distribution and initial facet joint gap thickness had little influence on the lumbar range of motion in any direction, whereas the mean contact pressure, total contact force, and total contact area predictions were altered considerably. The data indicate that range of motion predictions alone are insufficient to establish model validation intended to predict mechanical contact parameters. These data also emphasize the need for the careful consideration of the initial facet joint gap thickness with respect to the spinal condition being studied.     

Correlation of thoracic and lumbar vertebral failure loads with in situ vs. ex situ dual energy X-ray absorptiometry

In this study we explore the hypothesis that estimates of failure loads in the thoracic spine by lumbar dual energy X-ray absorptiometry (DXA) are compromised of skeletal heterogeneity throughout the spine and artifacts of spinal DXA. We studied the correlation between mechanical failure loads of thoracic and lumbar vertebrae, and that of in situ vs. ex situ lumbar DXA with thoracic and lumbar fracture loads, respectively. One hundred and nineteen subjects (76 female, age 82+/-9yr; 43 male, age 77+/-11yr) were examined under in situ conditions (anterior-posterior direction), the scans being repeated ex situ (lateral projection) in 68 cases. The failure loads of thoracic vertebrae (T) 6 and 10, and lumbar vertebra (L) 3 were determined in axial compression, using a functional 3-segment unit. The correlation between thoracic failure loads (T6 vs. T10) was significantly (p<0.01) higher (r=0.85) than those between thoracic and lumbar vertebrae (r=0.68 and 0.61, respectively). Lateral ex situ DXA displayed a significantly higher correlation (p<0.05) with lumbar vertebral fracture loads than in situ anterior-posterior DXA (r=0.85 vs. 0.71), but the correlation of thoracic failure loads with lateral ex situ lumbar DXA was similar to that obtained in situ in anterior-posterior direction (r=0.69 vs. 0.69 for T10, and r=0.61 vs. 0.65 for T6). The correlation between fracture loads of different spinal segments, and between DXA and failure loads was not significantly different between men and women. The results demonstrate a substantial heterogeneity of mechanical competence throughout the spine in elderly individuals. Because of the high incidence of fractures in the thoracic spine, these findings suggest that, clinically, lateral DXA involves no relevant advantage over anterior-posterior measurements of the lumbar spine.     

Regression model-based real-time markerless tumor tracking with fluoroscopic images for hepatocellular carcinoma

PurposeWe have developed a new method to track tumor position using fluoroscopic images, and evaluated it using hepatocellular carcinoma case data.MethodsOur method consists of a training stage and a tracking stage. In the training stage, the model data for the positional relationship between the diaphragm and the tumor are calculated using four-dimensional computed tomography (4DCT) data. The diaphragm is detected along a straight line, which was chosen to avoid 4DCT artifact. In the tracking stage, the tumor position on the fluoroscopic images is calculated by applying the model to the diaphragm. Using data from seven liver cases, we evaluated four metrics: diaphragm edge detection error, modeling error, patient setup error, and tumor tracking error. We measured tumor tracking error for the 15 fluoroscopic sequences from the cases and recorded the computation time.ResultsThe mean positional error in diaphragm tracking was 0.57 ± 0.62 mm. The mean positional error in tumor tracking in three-dimensional (3D) space was 0.63 ± 0.30 mm by modeling error, and 0.81–2.37 mm with 1–2 mm setup error. The mean positional error in tumor tracking in the fluoroscopy sequences was 1.30 ± 0.54 mm and the mean computation time was 69.0 ± 4.6 ms and 23.2 ± 1.3 ms per frame for the training and tracking stages, respectively.ConclusionsOur markerless tracking method successfully estimated tumor positions. We believe our results will be useful in increasing treatment accuracy for liver cases.     

Vertebrae numbers of the early hominid lumbar spine

General doctrine holds that early hominids possessed a long lumbar spine with six segments. This is mainly based on Robinson's (1972) interpretation of a single partial Australopithecus africanus skeleton, Sts 14, from Sterkfontein, South Africa. As its sixth last presacral vertebra exhibits both thoracic and lumbar characteristics, current definitions of lumbar vertebrae and lumbar ribs are discussed in the present study. A re-analysis of its entire preserved vertebral column and comparison with Stw 431, another partial A. africanus skeleton from Sterkfontein, and the Homo erectus skeleton KNM-WT 15000 from Nariokotome, Kenya, did not provide strong evidence for the presence of six lumbar vertebrae in either of these early hominids. Thus, in Sts 14 the sixth last presacral vertebra has on one side a movable rib. In Stw 431, the corresponding vertebra shows indications for a rib facet. In KNM-WT, 15000 the same element is very fragmentary, but the neighbouring vertebrae do not support the view that it is L1. Although in all three fossils the transitional vertebra at which the articular facets change orientation seems to be at Th11, this is equal to a large percentage of modern humans. Indeed, a modal number of five lumbar vertebrae, as in modern humans, is more compatible with evolutionary principles. For example, six lumbar vertebrae would require repetitive shortening and lengthening not only of the lumbar, but also of the entire precaudal spine. Furthermore, six lumbar vertebrae are claimed to be biomechanically advantageous for early hominid bipedalism, yet an explanation is lacking as to why the lumbar region should have shortened in later humans. All this raises doubts about previous conclusions for the presence of six lumbar vertebrae in early hominids. The most parsimonious explanation is that they did not differ from modern humans in the segmentation of the vertebral column.     

Does the correlation between schmorl's nodes and vertebral morphology extend into the lumbar spine?

Schmorl's nodes are depressions on vertebrae due to herniation of the nucleus pulposus of the intervertebral disc into the vertebral body. This study provides an extension of our previous study which analyzed the shape of the lower thoracic spine and found that vertebral morphology was associated with the presence of Schmorl's nodes. Ninety adult individuals from the late Medieval site of Fishergate House, York, and the Post‐Medieval site of Coach Lane, North Shields, Tyne and Wear, England, were analysed using 2D geometric morphometrics to identify possible relationships between vertebral morphology and Schmorl's nodes at the thoraco‐lumbar junction and in the lumbar spine. A significant correlation was found between vertebral shape and the presence of Schmorl's nodes in the twelfth thoracic vertebrae and the first to third lumbar vertebrae. The findings corroborate previous studies and suggest that vertebral shape may be an important factor in spinal health. It is hypothesized that the pedicle shape of affected vertebrae may not provide adequate structural support for the vertebral bodies, resulting in vertical disc herniation. Am J Phys Anthropol 157:526–534, 2015. © 2015 Wiley Periodicals, Inc.     

Functional implications of variation in lumbar vertebral count among hominins

As early as the 1970s, Robinson defined lumbar vertebrae according to their zygapophyseal orientation. He identified six lumbar elements in fossil Sts 14 Australopithecus africanus, one more than is commonly present in modern humans. It is now generally inferred that the modal number of lumbar vertebrae for australopiths and early Homo was six, from which the mode of five in later Homo is derived. The two central questions this study investigates are (1) to what extent do differences in human lumbar vertebral count affect lordotic shape and lumbar function, and (2) what does lumbar number variation imply about lumbar spine function in early hominins? To address these questions, I first outline a biomechanical model of lumbar number effect on lordotic function. I then identify relevant morphological differences in the human modal and extra-modal variants, which I use to test the model. These tests permit evaluation of the human L6 variant as a model for reconstructing early hominin modal number and spine function. Application of the biomechanical model in reconstructing australopith/early Homo lumbar spines highlights shared principles of Euler column strength and sagittal spine flexibility among early and modern hominins. Within modern humans, the extra-modal L6 variant has an extended series of three cranially positioned kyphotic vertebrae and strongly oblique zygapophyseal facets at the last lumbar level. Although they share the same radius and length of lumbar curvature, the L6 variant differs functionally from the L5 mode in its expanded range of sagittal flexion/extension and enhanced resistance to shear. Given the modal number of six lumbar vertebrae in australopiths and early Homo, lumbar spine mobility and strength would have been key properties of vertebral function in early bipeds whose upper and lower body segments were coupled by close approximation of the thorax and iliac crests.     

Quantitative anatomy of the lumbar musculature

This paper describes the anatomy of the musculature crossing the lumbar spine in a standardized form to provide data generally suitable for static biomechanical analyses of muscle and spinal forces. The muscular anatomy from several sources was quantified and transformed to the mean bony anatomy of four young healthy adults measured from standing stereo-radiographs. The origins, insertions and physiological cross-sectional area (PCSA) of 180 muscle slips which act on the lumbar spine are given relative to the bony anatomy defined by the locations of 12 thoracic and five lumbar vertebrae, and the sacrum, and the shape and positions of the 24 ribs. The broad oblique abdominal muscles are each represented by six vectors and an appropriate proportion of the total PCSA was assigned to each to represent the muscle biomechanics.     

DR全脊柱成像技术应用探讨

目的探讨DR全脊柱成像技术在脊柱侧弯畸形诊断及测量中的价值,研究不同检查方法对成像质量及测量精度的影响。方法利用PHIIAPS公司Digital Diagnost DR系统和具有全脊柱拼接功能的SUN后处理工作站,对分次所摄的脊柱正位、侧位、左、右侧屈位影像进行拼接处理后作必要的测量,并打印成像。结果经上述处理后的全脊柱正位、侧位及左、右侧屈位均能将颈、胸、腰、骶椎完整联接,显示在1张14＋17的X光片上,其测量的数据真实可靠。结论DR全脊柱成像技术克服了以往颈、胸、腰、骶椎单独成像后测量上的误差,为临床手术提供了更精确的数据。     

能谱CT在去除腰椎术后金属伪影中的应用进展

多层计算机断层扫描(Multi-slice computed tomography,MSCT)具有时间及空间分辨率高的特点已经被广泛应用于脊柱术后评价,然而,因为术中植入金属植入物,CT图像由于其成像原理的限制,其图像质量受到金属植入物的严重影响,金属伪影大多来源于量子噪声、散射线和射束硬化效应,降低图像的对比度,使细微解剖结构显示不清,从而使得感兴趣区域的检测能力降低,其图像质量明显下降,甚至使临床医师做出错误的诊断,因此,为了做出更准确的腰椎术后并发症诊断和术后评估,使金属伪影最小化成为关键。能谱CT及其扫描及重建技术可以有效的去除腰椎术后产生的金属伪影,改善图像质量,本文能谱CT在腰椎术后金属伪影去除方面的应用进行综述。     

赤峰兴隆沟遗址人类椎骨疾病的生物考古学研究

对赤峰兴隆沟遗址第一地点出土的人类椎骨上椎体的退行性关节病(DJD)和黄韧带骨化(OLF)进行了生物考古学研究,制订了这两种病症患病程度的分级记录标准,并对患病率和患病程度均值进行了统计分析。对两性的对比发现DJD的差异主要体现在男性颈椎患病较多而女性腰椎患病较多,OLF的主要差异则是男性胸椎患病率显著高于女性。通过与对比组的比较,发现兴隆沟组腰椎、颈椎DJD患病率相对较高,并再次验证了两性患病情况存在的明显差异;兴隆沟组OLF患病率与对比组也有显著差异。兴隆沟组的椎骨患病情况不一定与生业模式有关,只反映了脊柱承受负荷的情况,其差异暗示行为模式上的差异。建议开展其他方面研究以详细了解其行为模式,同时要重视建立统一的古病理研究标准的重要性。     

Mechanical variables affecting balloon kyphoplasty outcome--a finite element study

It is still unclear how a vertebral fracture should be stabilised and strengthened without endangering the remaining intact bone of the augmented vertebra or the adjacent vertebrae. Numerical modelling may provide insight. To date, however, few finite element (FE) spine models have been developed which are both multi-segmental and capture a more complete anatomy of the vertebrae. A 3-D, two-functional unit, CT-based, lumbar spine, FE model was developed and used to predict load transfer and likelihood of fracture following balloon kyphoplasty. The fractured anterior wall and injected cement were modelled in a two-functional spinal unit model with osteoporotic bone properties. Parameters investigated included: cement stiffness, cement volume and height restoration. Models were assessed based on stresses and a user-defined fracture-predicting field. Augmentation altered the stress distribution; shielding was dependent on positioning of the cement; and fracture algorithm found incomplete height restoration to increase the likelihood of fracture, particularly in adjacent vertebrae.     

Precise particle tracking against a complicated background: polynomial fitting with Gaussian weight

We present a new particle tracking software algorithm designed to accurately track the motion of low-contrast particles against a background with large variations in light levels. The method is based on a polynomial fit of the intensity around each feature point, weighted by a Gaussian function of the distance from the centre, and is especially suitable for tracking endogeneous particles in the cell, imaged with bright field, phase contrast or fluorescence optical microscopy. Furthermore, the method can simultaneously track particles of all different sizes, and allows significant freedom in their shape. The algorithm is evaluated using the quantitative measures of accuracy and precision of previous authors, using simulated images at variable signal-to-noise ratios. To these we add new tests: the error due to a non-uniform background, and the error due to two particles approaching each other. Finally the tracking of particles in real cell images is demonstrated. The method is made freely available for non-commercial use as a software package with a graphical user-interface, which can be run within the Matlab programming environment.     

Biomechanical design of load simulation in multiple spinal segments]

Multisegmental biomechanical studies on the lumbar spine are steadily increasing in importance. Only in this way can we acquire knowledge about the physiological behaviour of the entire lumbar spine. Furthermore, these studies allow us to analyse in vitro the biomechanics of manipulated lumbar spines after various surgical operations on the spine. A load simulator was developed to investigate multisegmental lumbar spine mobility, and its function was investigated in an initial study on 19 fresh--frozen specimens of human lumbar spine. After x-ray examination and determination of the bone mineral density, the specimens were loaded up to 10 Nm in the automatic electromechanical loading system under flexion/extension, lateral bending and axial rotation. An ultrasound-based motion analysis system was used to measure the displacements of the vertebrae involved.