微阵列图像的自动识别方法

微阵列图像具有高通量的特性,广泛应用于生物信息的研究中.微阵列图像的自动识别主要包括去除杂质噪声、网格定位、图像分割,还常包括倾斜图像的校正.现有的除噪方法主要有整体法、局部法、形态学滤波和小波变换.网格的自动定位主要是基于模版匹配或投影信号的信息,再利用图像的空间信息与像素强度进行分割,将图像的样点自动识别出来.分析了各部分不同算法的应用,并将其特点与结果加以综述.     

结合置信连接度的自适应模糊连接度的MRI图像中丘脑分割算法研究

目的立体定向神经外科手术中,丘脑及其子结构神经核团作为靶区,常被用于治疗癫痫和锥体外系疾病。利用计算机对丘脑神经核团进行分割,对于神经外科疾病的诊断与治疗具有重要的研究价值。为提高丘脑神经核团的分割精度,简化人工操作,减少人工干预,避免主观影响,本文提出一种结合置信连接度的自适应模糊连接度混合算法,用于分割丘脑结构。方法本文算法在基于模糊连接度的框架内增加图像梯度特征,采用自适应权重及自动选取感兴趣区域的方式,对10例人脑MRI图像数据的丘脑结构进行分割。结果实验结果与专家指导下的手工分割结果进行比较,并对两者之间的相似度进行量化比较。结果表明该算法在减少人工干预的同时保证了较高的准确率。结论结合置信连接度的自适应模糊连接度丘脑及其子结构分割算法在计算速度和精度上均优于传统模糊连接度算法。     

眼底图像中硬性渗出物检测算法

目的利用眼底图像中硬性渗出物(hard exudates,HE)的亮度与边缘特征,提出一种基于Canny边缘检测算法与形态学重构相结合的HE自动检测方法,以解决目前算法灵敏度低、检测结果中视盘和血管的干扰等问题,对糖尿病视网膜病变(diabetic retinopathy,DR)的自动筛查具有重要意义。方法检测算法包括4个步骤。步骤一,图像预处理,主要包括RGB通道选取、基于形态学的图像对比度增强。步骤二,视网膜图像关键结构的消除,利用基于Gabor滤波的血管分割方法,消除血管边缘对HE检测的影响。将本文视杯分割算法应用在眼底图像红色通道上实现视盘自动分割,消除视盘及其边缘对HE检测的影响。步骤三,利用改进的Canny边缘检测算法和形态学重构方法对HE进行提取。步骤四,基于形态学的图像后处理,消除眼底图像边缘部分假阳性区域。最后利用该算法测试公开数据库中的40幅图像(35幅HE病变图像,5幅正常图像)。结果该算法对基于病变的灵敏性(sensitivity,SE)和阳性预测值(positive predictive value,PPV)分别为93.18%和79.26%,基于图像的灵敏性、特异性(specificity,SP)和准确率(accuracy,ACC)分别为97.14%、80.00%和95.00%。结论与其他方法对比,基于Canny边缘检测算法与形态学重构相结合的HE自动检测算法具有较好的可行性。     

医学图像的自动调窗与分割

计算机辅助外科系统中,图像引导的手术导航系统是一种技术含量最高、发展最快的外科手术设备,在神经外科、骨科、耳鼻喉科等有着广阔的应用前景。目前,手术导航系统在图像调窗、分割、配准均采用的是手工方式,迫切需要提高图像处理的自动化程度。本文提出的自动调窗与分割算法,比较好地实现了医学图像的自动调窗和自动分割功能:(1)通过对大量MRI和CT图像的灰度直方图分析,根据这一类图像的共性,给出相应的调窗算法,实验表明,该算法绝大部分MRI和CT图像的自动调窗效果接近于最佳;(2 )本文给出的种子生长分割方法,是基于灰度连通性原理,只需点击病灶,就可以将病灶及边缘准确地分割出来,并可以动态、实时地控制分割的效果,只要机器的内存允许,可以直接对三维图像进行三维分割。测试表明,该功能缩短和降低手术计划的时间和难度,提高导航手术的效率。     

基于自适应标记分水岭算法的肝脏CT图像自动分割

目的为减少人工交互提出了基于自适应标记分水岭的CT系列图像肝脏区域自动分割算法。方法首先对图像进行形态学重构运算以平滑图像,然后计算多尺度形态学梯度,同时提出利用梯度图像非零的局部极小值点的均值进行自适应标记提取,以避免分水岭的过分割和欠分割,再结合肝脏为最大的实质性脏器和相邻图像的相似性实现CT系列图像的肝区自动分割。结果该算法能自动、快速地提取CT系列图像中的肝脏区域。结论分水岭算法能准确定位区域的边缘,通过选择合适的阈值对梯度图像进行标记以抑制分水岭的过分割,实现医学图像中感兴趣区域的自动分割。     

免疫组化彩色细胞图像自动分割的研究

目的：免疫组化彩色细胞图像中阳性细胞的自动分割提取有着重要的临床意义。本文结合三种分割算法的特点，研究实现免疫组化彩色细胞图像的自动分割，提取图像中的阳性细胞。方法：（1）采用OTSU方法在灰度的基础上对免疫组化彩色细胞进行预分割，去除无关背景。（2）使用K-聚类算法，对彩色细胞图像进行彩色分类筛选出阳性细胞和阴性细胞，并对所得阳性细胞图像进行腐蚀，以获取阳性细胞图像的种子。（3）使用区域生长算法对阳性细胞种子进行区域增长。获取完整的阳性细胞图。结果：准确分割出图像中的阳性细胞。结论：该自动分割方法可用于后续的阳性细胞自动计数，辅助医生诊断疾病。     

基于人工生命的医学图像分割

医学图像中解剖结构和相关诊断信息的提取有着极为重要的意义，但目前的分割算法大都需要借助专家的干预和监控，寻求一种全自动分割的方法变得日益重要。人工生命的方法有助于人们了解生物学规律，并且在机器人、计算机图形学等方面得到了成功应用。主要介绍基于人工生命的方法在医学图像自动分割领域的初步应用。     

基于卷积神经网络和图像显著性的心脏CT图像分割

目的心脏医学影像中,感兴趣部分的提取与分割是诊断心脏病变部位的关键。由于心脏舒张、收缩以及血液的流动,心脏CT图像易出现弱边界、伪影,传统分割算法易产生过度分割的情况。为此,提出一种基于卷积神经网络和图像显著性的心脏CT图像分割方法。方法采用卷积神经网络对目标区域进行定位,滤除肋骨、肌肉等造影对比不明显部分,截取出感兴趣区域,结合感兴趣区域的对比度计算并提高感兴趣区域的心脏组织的显著值。通过获得的显著值图像截取心脏图像,并与区域生长算法的分割结果进行对比。最后使用泰州人民医院11例患者的影像数据对算法模型进行训练和测试,随机选择9例用于训练,剩余2例用于测试。结果所提算法模型在心底、心中、心尖3个心脏分段的分割正确率分别达到了92.79%、92.79%、94.11%,均优于基于区域生长的分割方法。结论基于卷积神经网络和图像显著性的分割方法能够准确获取心脏的外围轮廓,轮廓边缘更加平滑,完全能够满足CT图像序列的心脏全自动分割任务需求,分割后的图像更有利于医生对患者心脏健康状况和病变部位的观察。     

基于对称区域生长算法的超声医学图像分割方法

提出了一种基于对称区域生长算法的超声医学图像的分割方法。该方法分为三步。首先,通过采用自适应加权中值滤波抑制超声医学图像本身固有的Speckle噪声,然后从图像的第一行开始扫描整个图像,并应用生长准则进行区域的生长与合并,生长完成之后应用种子准则标定感兴趣区域,从而得到最后的分割结果。通过图像的分割实验确定了一套对于超声医学图像适用的生长和合并准则。对心脏B型超声医学图像分割的实验结果显示,该方法具有良好的性能。     

模糊神经网络在颅脑磁共振图像分割中的应用研究

由于颅脑磁共振图像分割的效果受制于解剖结构和成像过程引起的边缘模糊和噪声，本研究结合神经网络和模糊逻辑技术，提出一种基于模糊神经网络的颅脑磁共振图像的半自动分割算法。分割实验的结果表明，在同等条件下，模糊神经网络分割算法的收敛速度比BP神经网络分割算法快三倍以上；与最大似然法、模糊c均值聚类和BP神经网络分割算法相比，FNN分割算法的抗噪声和抗模糊能力更强。     

A Fast and Fully Automatic Method for Cerebrovascular Segmentation on Time-of-Flight (TOF) MRA Image

The precise three-dimensional (3-D) segmentation of cerebral vessels from magnetic resonance angiography (MRA) images is essential for the detection of cerebrovascular diseases (e.g., occlusion, aneurysm). The complex 3-D structure of cerebral vessels and the low contrast of thin vessels in MRA images make precise segmentation difficult. We present a fast, fully automatic segmentation algorithm based on statistical model analysis and improved curve evolution for extracting the 3-D cerebral vessels from a time-of-flight (TOF) MRA dataset. Cerebral vessels and other tissue (brain tissue, CSF, and bone) in TOF MRA dataset are modeled by Gaussian distribution and combination of Rayleigh with several Gaussian distributions separately. The region distribution combined with gradient information is used in edge-strength of curve evolution as one novel mode. This edge-strength function is able to determine the boundary of thin vessels with low contrast around brain tissue accurately and robustly. Moreover, a fast level set method is developed to implement the curve evolution to assure high efficiency of the cerebrovascular segmentation. Quantitative comparisons with 10 sets of manual segmentation results showed that the average volume sensitivity, the average branch sensitivity, and average mean absolute distance error are 93.6%, 95.98%, and 0.333 mm, respectively. By applying the algorithm to 200 clinical datasets from three hospitals, it is demonstrated that the proposed algorithm can provide good quality segmentation capable of extracting a vessel with a one-voxel diameter in less than 2 min. Its accuracy and speed make this novel algorithm more suitable for a clinical computer-aided diagnosis system.     

Semiautomated Quantitation of Carotid Artery Stenosis in Gadolinium-Bolus Magnetic Resonance Angiography

Magnetic resonance angiography (MRA) has become the standard method for evaluation of carotid occlusive disease. Fast imaging methods combined with bolus intravenous injection of gadolinium contrast have improved the quality of these images. Nevertheless, the gold standard for evaluation was based on projection arterial angiography. The properties of these images are rather different. Whereas most previous evaluations of MRA have used visual assessment of images, we evaluate an algorithm in which a computer algorithm plays the primary role in defining arterial lumen margins, hence, disease. The accuracy of this semiautomated algorithm is shown to compare favorably with gold-standard arteriography in a series of 50 patients.     

Arterial Head Vascularization Cartographies of Normal Metencephalic Dogs Using Magnetic Resonance Angiography

The aim of our study was to establish head arterial cartographies—useful for the diagnosis of brain diseases leading to cerebral vascular modifications—by means of magnetic resonance angiography (MRA). Casts of the arterial vascular brain system were used to corroborate the MRA results as they can be easily rotated in nonvirtual three‐dimensions and give an accurate view of the arteries calibre and origin. Two types of 3T MRA images were used: three‐dimensional fast low‐angle shot (3D‐FLASH) acquisition sequenced every 20 s, paired with injection of a paramagnetic contrast medium, and three‐dimensional time‐of‐flight (3D‐TOF) acquisition sequenced every 300 s. 3D‐FLASH acquisition gives very accurate images of the cerebral arteries and veins, but must be used with care in debilitated animals. 3D‐TOF acquisition is less accurate and gives only images of the main cerebral arteries without showing the venous system. It is, however, a viable diagnostic method for monitoring vascular lesions (e.g., cerebral hemorrhages). Anat Rec, 2011. © 2011 Wiley‐Liss, Inc.     

Combined use of T1-weighted MRI and MRA for stereotaxic lesioning of the nonhuman primate brain: application to the rhinal cortex

 Stereotaxic brain lesioning is widely used to develop experimental models of human brain disease in the nonhuman primate. To avoid intraoperative vascular complications such as intracranial hemorrhage, we developed a methodology that is easy to implement. This method combines T1-weighted magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA). This technique is applied to produce bilateral neurotoxic lesions of the rhinal cortex, a structure located medially in the temporal lobe, in eight baboons (including five sham-operated animals with needle descents but no ibotenic acid injection). Two other baboons were lesioned before the MRA technology was available. The MRA sequence (two-dimensional time-of-flight, axial acquisition) was used to localize the blood vessels in the needle trajectories, i.e., the highly vascularized sylvian fissure and temporal gyri. The vessel coordinates were transposed onto the coronal MRI-T1 images, onto which the injection sites were determined and the planned needle tracks drawn. In the eight baboons that had MRA, 26.8% of these needle tracks had to be slightly displaced because of the presence of blood vessels. The stereotaxic coordinates of the final target sites were then calculated with respect to six skull landmarks that also served as a reference during surgery. No intracranial hemorrhage occurred in any of the eight baboons in which MRA was performed, in contrast to one of the two baboons not studied with MRA. The histological analysis showed a good extent of the rhinal lesions in all lesioned animals, with minimal damage to areas other than those that were targeted. Thus, combined use of MRI-T1 and MRA proved to be reliable in reducing vascular complications, affording new advances for stereotaxic surgery in nonhuman primates. Received: 16 June 1998 / Accepted: 13 November 1998     

Simplified skipped phase encoding and edge deghosting (SPEED) for imaging sparse objects with applications to MRA

The fast imaging method named skipped phase encoding and edge deghosting (SPEED) has been demonstrated to reduce scan time considerably with typical magnetic resonance imaging data. In this work, SPEED is simplified with improved efficiency to accelerate the scan of sparse objects; we refer to this method as S-SPEED. S-SPEED partially samples k-space into two interleaved data sets, each with the same skip size of N but a different relative shift in phase encoding. The sampled data are then Fourier transformed into two ghosted images with N aliasing ghosts. Given the sparseness of signal distribution, the ghosted images are simply modeled with a single-layer structure, analogous to that used in maximum-intensity projection. With an algorithm based on a least-square-error solution, a deghosted image is solved, and a residual map is output for quality control. S-SPEED can be generalized to include more layers with additional acquisitions for refined results. Without differential filtering and full central k-space sampling, S-SPEED reduces scan time further and achieves more straightforward reconstruction, as compared with SPEED. In this work, S-SPEED is applied to accelerate magnetic resonance angiography (MRA) by taking advantage of the sparse nature of MRA data. With sparse phantom data and in vivo phase contrast MRA data, S-SPEED is demonstrated to achieve satisfactory results with an acceleration factor of 5.5 using a single coil.     

Serum complement levels in patients with rheumatoid arthritis and vasculitis

It is known that serum complement levels are decreased in patients with rheumatoid arthritis associated with vasculitis, also called malignant rheumatoid arthritis (MRA). The complement profiles in patients with MRA were compared with those in uncomplicated rheumatoid arthritis (RA) and in those with systemic lupus erythematosus (SLE). In MRA patients, serum CH50, C4 and C3 levels were all decreased as in SLE patients; and C3d, an activation product of C3, and the C3d/C3 ratio as a C3 activation index were increased. Serum B level was also increased as in RA patients, but AH50 level as the haemolytic activity of the alternative pathway was within the normal range. The regulatory proteins, H and I, were elevated in sera of MRA patients. Circulating immune complexes (CIC) detected by a solid phase C1q binding method were increased, whereas the serum activity to solubilize a performed immune complex (CRA) was decreased in MRA sera. It was suggested that increased consumption of complements, mainly through the classical pathway may be responsible for the complement profiles in MRA patients: a markedly consumed classical pathway and relatively preserved alternative pathway.     

Radial 3D TOF和常规3D TOF在颈动脉成像中的对比

目的:探讨Radial 3D Time of Flight（TOF）序列在颈动脉磁共振血管成像中的应用。方法:选取行颈动脉常规3D TOF和Radial 3D TOF的36例患者的图像资料，采用配对样本t检验对比两个序列颈动脉的信噪比、对比噪声比，采用非参数配对Wilcoxon符号秩检验对比两个序列颈动脉血管的显示效果（主观评分）。结果:Radial 3D TOF的信噪比和对比噪声比均高于常规3D TOF（t=25.7、14.2, P值均小于0.05）；血管的显示效果也优于常规3D TOF（z=-4.36, P<0.05）。结论:相比于常规3D TOF，Radial 3D TOF可以获得更好的颈动脉图像质量。     

Time-of-flight-, phase contrast and contrast enhanced magnetic resonance angiography for pre-interventional determination of aneurysm size,configuration, and neck morphology in an aneurysm model in rabbits

We describe three different magnetic resonance (MR)-angiography techniques to evaluate aneurysm size, configuration, and neck morphology of experimentally created aneurysms in a rabbit model. In five New Zealand White rabbits an aneurysm was created by endovascular occlusion of the right common carotid artery (CCA) using a pliable balloon and subsequent endoluminal incubation of elastase within the proximal CCA above the balloon and distal ligation of the vessel. In all animals, time-of-flight (TOF), phase contrast and contrast enhanced (CE) MR angiographies (MRA) were performed and compared to conventional digital subtraction angiography results. We found, that aneurysms are best visualized employing CE MRA, however, neck morphology was also found to demonstrate interpretable results when evaluating the axial source data of the TOF MRA. The animal model we used can be employed for testing endovascular devices such as new coil material, or covered stents. The described MRA techniques might then be helpful for pre-interventional planning and maybe even for the follow-up of the thus treated aneurysms.     

Computed tomography and magnetic resonance angiography in the evaluation of aberrant origin of the external carotid artery branches

Aim of our study was to evaluate the prevalence of aberrant origin of the branches of the external carotid artery (ECA) in 97 patients by computed tomography (CTA) and magnetic resonance angiography (MRA) and to compare the accuracy of these two techniques in the visualization of the ECA system. All patients underwent CTA and MRA examination of the head and neck. Multiplanar and volumetric reformations were obtained in all cases. For each set of images, the presence of aberrant origin of the branches of the external carotid artery was investigated. MRA and CTA images of each patient were compared to define their information content. Anatomical anomalies were found in 88 heminecks, with a prevalence of 53.3%. In the 61 patients in whom the CTA was performed before the MRA, the latter method showed only 92% of abnormalities detected at the first examination; in the 36 patients in whom MRA was performed first, CTA identified all of the anomalies highlighted by the former, adding 12 new. Knowledge of the anomalies of origin of the ECA branches is essential for the head and neck surgeon; the high prevalence of anomalies found in our series as in the previous studies indicates the opportunity to perform a CTA or a MRA of the head and neck before any surgical or interventional procedure. CTA is the method of choice in the evaluation of anomalies of origin of the branches of the ECA and in the definition of their course.     

Application of magnetic resonance imaging and magnetic resonance angiography as diagnostic measures for the first attack of suspected cerebrovascular diseases in Korea

Purpose

No precise data are available showing how magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) can be applied to diagnosis for the first attack of a suspected cerebrovascular disease in Korea. The purpose of this study was to evaluate the application level of MRI and MRA as diagnostic tools and the related factors to the use of these techniques.

Materials and Methods

This study used the health benefit claim data of 89,890 patients who were hospitalized for the first time due to suspected cerebrovascular disease in 2007 without having visited medical institutions as an outpatient or inpatient from 2003 to 2006.

Results

Of the 89,890 cases, 28.4% took both MRI and MRA, 10.7% took only MRI and 6.9% took only MRA. The related factors identified in the multivariate logistic regression analysis were gender, type of insurance, type of medical institution, type of department, duration of hospitalization, and type of disease.

Conclusion

This study showed that the application level of MRI and MRA as diagnostic measures for the first attack of a suspected cerebrovascular diseases varied depending on several factors. It is necessary to study more accurate levels of computerized tomography (CT), computerized tomography angiography (CTA), MRI or MRA as measures to diagnose a first attack of suspected cerebrovascular disease.