Lung metastases detection in CT images using 3D template matching

The aim of this study is to demonstrate a novel, fully automatic computer detection method applicable to metastatic tumors to the lung with a diameter of 4-20 mm in high-risk patients using typical computed tomography (CT) scans of the chest. Three-dimensional (3D) spherical tumor appearance models (templates) of various sizes were created to match representative CT imaging parameters and to incorporate partial volume effects. Taking into account the variability in the location of CT sampling planes cut through the spherical models, three offsetting template models were created for each appearance model size. Lung volumes were automatically extracted from computed tomography images and the correlation coefficients between the subregions around each voxel in the lung volume and the set of appearance models were calculated using a fast frequency domain algorithm. To determine optimal parameters for the templates, simulated tumors of varying sizes and eccentricities were generated and superposed onto a representative human chest image dataset. The method was applied to real image sets from 12 patients with known metastatic disease to the lung. A total of 752 slices and 47 identifiable tumors were studied. Spherical templates of three sizes (6, 8, and 10 mm in diameter) were used on the patient image sets; all 47 true tumors were detected with the inclusion of only 21 false positives. This study demonstrates that an automatic and straightforward 3D template-matching method, without any complex training or postprocessing, can be used to detect small lung metastases quickly and reliably in the clinical setting.     

Lung nodule diagnosis using 3D template matching

In this paper, to utilize the third dimension of Computed Tomography, regions of interest (ROI) slices were combined to form 3D ROI image and a 3D template was determined to find the structures with similar properties of nodules. Convolution of 3D ROI image with the proposed template strengthens the shapes similar to the template and weakens the other ones. False-positive (FP) per nodule and per slice versus diagnosis sensitivity were obtained. The Computer Aided Diagnosis system achieved 100% sensitivity with 0.83 FP per nodule and 0.46 FP per slice, when the nodule thickness was greater than or equal to 5.625 mm.     

Breast MR segmentation and lesion detection with cellular neural networks and 3D template matching

A novel fully automated system is introduced to facilitate lesion detection in dynamic contrast-enhanced, magnetic resonance mammography (DCE-MRM). The system extracts breast regions from pre-contrast images using a cellular neural network, generates normalized maximum intensity-time ratio (nMITR) maps and performs 3D template matching with three layers of 12x12 cells to detect lesions. A breast is considered to be properly segmented when relative overlap >0.85 and misclassification rate <0.10. Sensitivity, false-positive rate per slice and per lesion are used to assess detection performance. The system was tested with a dataset of 2064 breast MR images (344slicesx6 acquisitions over time) from 19 women containing 39 marked lesions. Ninety-seven percent of the breasts were segmented properly and all the lesions were detected correctly (detection sensitivity=100%), however, there were some false-positive detections (31%/lesion, 10%/slice).     

Mutual information-based template matching scheme for detection of breast masses: from mammography to digital breast tomosynthesis

Development of a computational decision aid for a new medical imaging modality typically is a long and complicated process. It consists of collecting data in the form of images and annotations, development of image processing and pattern recognition algorithms for analysis of the new images and finally testing of the resulting system. Since new imaging modalities are developed more rapidly than ever before, any effort for decreasing the time and cost of this development process could result in maximizing the benefit of the new imaging modality to patients by making the computer aids quickly available to radiologists that interpret the images. In this paper, we make a step in this direction and investigate the possibility of translating the knowledge about the detection problem from one imaging modality to another. Specifically, we present a computer-aided detection (CAD) system for mammographic masses that uses a mutual information-based template matching scheme with intelligently selected templates. We presented principles of template matching with mutual information for mammography before. In this paper, we present an implementation of those principles in a complete computer-aided detection system. The proposed system, through an automatic optimization process, chooses the most useful templates (mammographic regions of interest) using a large database of previously collected and annotated mammograms. Through this process, the knowledge about the task of detecting masses in mammograms is incorporated in the system. Then, we evaluate whether our system developed for screen-film mammograms can be successfully applied not only to other mammograms but also to digital breast tomosynthesis (DBT) reconstructed slices without adding any DBT cases for training. Our rationale is that since mutual information is known to be a robust inter-modality image similarity measure, it has high potential of transferring knowledge between modalities in the context of the mass detection task. Experimental evaluation of the system on mammograms showed competitive performance compared to other mammography CAD systems recently published in the literature. When the system was applied “as-is” to DBT, its performance was notably worse than that for mammograms. However, with a simple additional preprocessing step, the performance of the system reached levels similar to that obtained for mammograms. In conclusion, the presented CAD system not only performed competitively on screen-film mammograms but it also performed robustly on DBT showing that direct transfer of knowledge across breast imaging modalities for mass detection is in fact possible.     

Analysis of diffusion tensor magnetic resonance imaging data using principal component analysis

An analysis method for diffusion tensor (DT) magnetic resonance imaging data is described, which, contrary to the standard method (multivariate fitting), does not require a specific functional model for diffusion-weighted (DW) signals. The method uses principal component analysis (PCA) under the assumption of a single fibre per pixel. PCA and the standard method were compared using simulations and human brain data. The two methods were equivalent in determining fibre orientation. PCA-derived fractional anisotropy and DT relative anisotropy had similar signal-to-noise ratio (SNR) and dependence on fibre shape. PCA-derived mean diffusivity had similar SNR to the respective DT scalar, and it depended on fibre anisotropy. Appropriate scaling of the PCA measures resulted in very good agreement between PCA and DT maps. In conclusion, the assumption of a specific functional model for DW signals is not necessary for characterization of anisotropic diffusion in a single fibre.     

Computerized detection of breast masses in digitized mammograms

We propose a system to detect malignant masses on mammograms. We investigated the behavior of an iris filter at different scales. After iris filter was applied, suspicious regions were segmented by means of an adaptive threshold. Suspected regions were characterized with features based on the iris filter output and, gray level, texture, contour-related, and morphological features extracted from the image. A backpropagation neural network classifier was trained to reduce the number of false positives. The system was developed and evaluated with two completely independent data sets. Results for a test set of 66 malignant and 49 normal cases, evaluated with free-response receiver operating characteristic analysis, yielded a sensitivity of 88% and 94% at 1.02 false positives per image for lesion-based and case-based evaluation, respectively. Results suggest that the proposed method could help radiologists as a second reader in mammographic screening.     

猫视神经连续切片的计算机三维重建

目的制备猫视神经的连续切片,选择合适的染色方法 ,寻找合适的配准方法 ,经图像采集及初步处理后利用计算机的三维重建技术,获得可视化的猫视神经三维结构。方法利用石蜡切片技术的制备猫视神经连续切片,采用特殊的神经三色染色法进行切片染色。利用数码显微镜及Motic Images Advanced 3.0和Motic Images Assembly 1.0软件对连续的猫视神经石蜡切片进行拍照,获得猫视神经连续切片的原始图像,将原始图像在GIMP软件中经过调整亮度对比度、初步图像配准、提取神经束膜图像信息后在Amira软件下,经过计算机在手工配准的基础上进一步自动配准,把二维的切片图像重建成三维的可视化图像。结果实验所得经三色染色的连续切片的原始图像上神经束膜能较好的与周围的组织区分;利用Amira三维重建软件,重建得到猫视神经三维结构图像,重建后的猫视神经结构三维图像可进行任意的切割、旋转、回复、拍照等操作。结论实验证明在不利用标志线的条件下,通过制备视神经连续切片,经过染色拍照,手工图像配准后利用计算机自动配准技术对视神经进行三维重建是完全可行的。     

Computerized detection of breast cancer with artificial intelligence and thermograms

This paper shows the concurrent use of thermography and artificial neural networks (ANN) for the diagnosis of breast cancer, a disease that is growing in prominence in women all over the world. It has been reported that breast thermography itself could detect breast cancer up to 10 years earlier than the conventional golden methods such as mammography, in particular in the younger patient. However, the accuracy of thermography is dependent on many factors such as the symmetry of the breasts' temperature and temperature stability. A woman's body temperature is known to be stable in certain periods after menstruation and it was found that the accuracy of thermography in women whose thermal images are taken in a suitable period (5th - 12th and 21st day of menstruation) is higher (80%) than the total population of patients (73%). The stability of the body temperature will depend on physiological state. This paper examines the use of ANN to complement the infrared heat radiating from the surface of the body with other physiological data. Four backpropagation neural networks were developed and trained using the results from the Singapore General Hospital patients' physiological data and thermographs. Owing to the inaccuracies found in thermography and the low population size gathered for this project, the networks developed could only accurately diagnose about 61.54% of the breast cancer cases. Nevertheless, the basic neural network framework has been established and it has great potential for future development of an intelligent breast cancer diagnosis system. This would be especially useful to the teenagers and young adults who are unsuitable for mammography at a young age. An intelligent breast thermography-neural network will be able to give an accurate diagnosis of breast cancer and can make a positive impact on breast disease detection.     

Computerized detection of breast cancer with artificial intelligence and thermograms

This paper shows the concurrent use of thermography and artificial neural networks (ANN) for the diagnosis of breast cancer, a disease that is growing in prominence in women all over the world. It has been reported that breast thermography itself could detect breast cancer up to 10 years earlier than the conventional golden methods such as mammography, in particular in the younger patient. However, the accuracy of thermography is dependent on many factors such as the symmetry of the breasts' temperature and temperature stability. A woman's body temperature is known to be stable in certain periods after menstruation and it was found that the accuracy of thermography in women whose thermal images are taken in a suitable period (5th - 12th and 21st day of menstruation) is higher (80%) than the total population of patients (73%). The stability of the body temperature will depend on physiological state. This paper examines the use of ANN to complement the infrared heat radiating from the surface of the body with other physiological data. Four backpropagation neural networks were developed and trained using the results from the Singapore General Hospital patients' physiological data and thermographs. Owing to the inaccuracies found in thermography and the low population size gathered for this project, the networks developed could only accurately diagnose about 61.54% of the breast cancer cases. Nevertheless, the basic neural network framework has been established and it has great potential for future development of an intelligent breast cancer diagnosis system. This would be especially useful to the teenagers and young adults who are unsuitable for mammography at a young age. An intelligent breast thermography-neural network will be able to give an accurate diagnosis of breast cancer and can make a positive impact on breast disease detection.     

Interpretation of laboratory results using multidimensional scaling and principal component analysis

Principal component analysis (PCA) and multidimensional scaling (MDS) are a set of mathematical techniques which uncover the underlying structure of data by examining the relationships between variables. Both MDS and PCA use proximity measures such as correlation coefficients or Euclidean distances to generate a spatial configuration (map) of points where distances between points reflect the relationship between individuals with their underlying set of data. Multidimensional scaling, when compared to PCA, gives more readily interpretable solutions of lower dimensionality and does not depend on the assumption of a linear relationship between variables. Both MDS and PCA were applied to electrolyte profiles of patients with acute renal failure and patients without apparent disease. The MDS was superior to PCA in separating renal patients from normal patients. The one-dimensional and two-dimensional solutions of MDS and PCA were compared.     

腹膜后脏器计算机三维可视化

目的:建立腹膜后重要脏器的计算机三维可视化模型。方法:应用中国数字化可视人体数据集,选取从。肾上腺顶部到肾底部的连续断面图像,在计算机上对肾等腹膜后脏器的断面图像轮廓进行数据分割,并对其行三维重建的立体显示。结果:重建出了肾上腺、肾、输尿管等脏器的三维可视化模型,该模型既可进行单个器官的显示,也可进行多个器官的分色显示,同时也可以任意放大缩小和任意角度旋转观察。结论:该腹膜后脏器的三维可视化模型展示了这些器官的三维空间结构,给临床影像诊断和外科手术提供了形态学参考。     

颈深筋膜及筋膜间隙的计算机三维重建

目的 :对颈深筋膜及深筋膜间隙进行计算机三维重建以显示其立体结构。方法 :用生物塑化技术制作薄层断面标本 ,在SGI工作站上 ,采用双线提取筋膜及间隙轮廓的方法对颈深筋膜及其间隙进行了三维重建。结果 :重建出咽后间隙、颈动脉间隙和内脏间隙及颈部的大血管和重要器官。重建结构均能单独显示、任意搭配显示或总体显示 ,可在三维空间位置上绕任意轴旋转任意角度。结构 :用双线法提取筋膜及间隙轮廓 ,能重建出筋膜间隙并能同时清楚显示间隙内的结构 ,为筋膜间隙的计算机三维重建提供了一种新的方法     

基于脉搏传感测值和主成分分析对精神疲劳状态的识别

背景：随着科技的进步,研究疲劳的客观手段越来越多,生理指标的介入使其成为医学、认知科学和心理学的研究热点。然而,对精神疲劳的检测目前仍缺乏客观的生理指标。 目的：为了评估精神疲劳状态,提出一种基于脉搏信号的精神疲劳状态识别新方法。 方法：用小波变换对脉搏信号消噪处理,提取脉搏信号功率谱峰值及对应频率、功率谱重心及重心频率特征量,对提取的特征量进行主成分分析,最后用改进的线性判别式分析法分类识别,主成分识别率达100%。 结果与结论：用脉搏信号特征的主成分进行精神疲劳状态识别,获得了满意的分类识别效果,该方法计算简单,稳定性好,识别率高,对精神疲劳状态的评估具有一定的可行性。     

Use of principal component factor analysis in the detection of carotid artery disease from Doppler ultrasound

Principal component factor analysis, a mathematical feature extraction technique, has been used to analyse the total information contained in the Doppler signal. In this study two patient groups have been investigated, normals and stenoses of less than 50%. The patients have been classified according to angiographic findings (patients with hypertension, migrane, heart disease, etc. havenot been excluded). The results from the principal component analysis technique have been compared with the more familiar A/B ratio based on the maximum frequency enevelope. Of the 25 normal vessel segments 20 were classified as normal by the A/B ratio technique and 22 by the principal component technique, while of the 19 abnormal vessels 13 were classified as abnormal by the A/B technique and 17 by the principal component analysis. Also the principal component analysis of the total Doppler signal was statistically superior to the A/B ratio in separating the two groups examined in this study.     

Computerized analysis of prostate lesions in the peripheral zone using dynamic contrast enhanced MRI

A novel automated computerized scheme has been developed for determining a likelihood measure of malignancy for cancer suspicious regions in the prostate based on dynamic contrast-enhanced magnetic resonance imaging (MRI) (DCE-MRI) images. Our database consisted of 34 consecutive patients with histologically proven adenocarcinoma in the peripheral zone of the prostate. Both carcinoma and non-malignant tissue were annotated in consensus on MR images by a radiologist and a researcher using whole mount step-section histopathology as standard of reference. The annotations were used as regions of interest (ROIs). A feature set comprising pharmacokinetic parameters and a T1 estimate was extracted from the ROIs to train a support vector machine as classifier. The output of the classifier was used as a measure of likelihood of malignancy. Diagnostic performance of the scheme was evaluated using the area under the ROC curve. The diagnostic accuracy obtained for differentiating prostate cancer from non-malignant disorders in the peripheral zone was 0.83 (0.75-0.92). This suggests that it is feasible to develop a computer aided diagnosis system capable of characterizing prostate cancer in the peripheral zone based on DCE-MRI.     

Automated breast mass detection in 3D reconstructed tomosynthesis volumes: a featureless approach

The purpose of this study was to propose and implement a computer aided detection (CADe) tool for breast tomosynthesis. This task was accomplished in two stages-a highly sensitive mass detector followed by a false positive (FP) reduction stage. Breast tomosynthesis data from 100 human subject cases were used, of which 25 subjects had one or more mass lesions and the rest were normal. For stage 1, filter parameters were optimized via a grid search. The CADe identified suspicious locations were reconstructed to yield 3D CADe volumes of interest. The first stage yielded a maximum sensitivity of 93% with 7.7 FPs/breast volume. Unlike traditional CADe algorithms in which the second stage FP reduction is done via feature extraction and analysis, instead information theory principles were used with mutual information as a similarity metric. Three schemes were proposed, all using leave-one-case-out cross validation sampling. The three schemes, A, B, and C, differed in the composition of their knowledge base of regions of interest (ROIs). Scheme A's knowledge base was comprised of all the mass and FP ROIs generated by the first stage of the algorithm. Scheme B had a knowledge base that contained information from mass ROIs and randomly extracted normal ROIs. Scheme C had information from three sources of information-masses, FPs, and normal ROIs. Also, performance was assessed as a function of the composition of the knowledge base in terms of the number of FP or normal ROIs needed by the system to reach optimal performance. The results indicated that the knowledge base needed no more than 20 times as many FPs and 30 times as many normal ROIs as masses to attain maximal performance. The best overall system performance was 85% sensitivity with 2.4 FPs per breast volume for scheme A, 3.6 FPs per breast volume for scheme B, and 3 FPs per breast volume for scheme C.     

Full data utilization in PVI using the 3D radon transform.

An algorithm is described for three-dimensional (3D) image reconstruction in positron volume imaging (PVI) using the inversion of the 3D radon transform (RT) for a truncated cylindrical detector geometry. A discrete version of the 3D RT is formed in a 3D histogram from the event-line coordinates of the detected events. The histogram entries represent the plane integrals of the activity in the field of view. Conventional inversion of the x-ray transform (XT) excludes oblique events in non-iterative reconstruction methods. by employing a spatially varying angular acceptance of events into the histogrammed plane integrals of the 3D RT, it is possible to include most of the detected oblique events in the reconstruction of the image using the standard 3D RT inversion formula. the oblique events are added to the histogram with a partial weight compared to those in complete (XT) projections. This single-pass reconstruction image has better statistical noise properties than images formed by RT inversion from complete XT projections, but only for some detector geometries is it significantly better. Monte Carlo simulations were used to study the statistical noise in images reconstructed using the new algorithm. The inherent difference in the axial versus the transaxial statistical noise in images reconstructed from truncated detectors is noted and is found to increase by including oblique events with this new algorithm.     

Statistical analysis of mitochondrial pathologies in childhood: identification of deficiencies using principal component analysis

Mitochondrial pathologies are a heterogeneous group of metabolic disorders that are frequently characterized by anomalies of oxidative phosphorylation, especially in the respiratory chain. The identification of these anomalies may involve many investigations, and biochemistry is a main tool. However, considering the whole set of biochemical data, the interpretation of the results by the traditionally used statistical methods remains complex and does not always lead to an unequivocal conclusion about the presence or absence of a respiratory chain defect. This arises from three main problems: (a) the absence of an a priori-defined control population, because the determination of the control values are derived from the whole set of investigated patients, (b) the small size of the population studied, (c) the large number of variables collected, each of which creates a wide variability. To cope with these problems, the principal component analysis (PCA) has been applied to the biochemical data obtained from 35 muscle biopsies of children suspected of having a mitochondrial disease. This analysis makes it possible for each respiratory chain complex to distinguish between different subsets within the whole population (normal, deficient, and, in between, borderline subgroups of patients) and to detect the most discriminating variables. PCA of the data of all complexes together showed that mitochondrial diseases in this population were mainly caused by multiple deficits in respiratory chain complexes. This analysis allows the definition of a new subgroup of newborns, which have high respiratory chain complex activity values. Our results show that the PCA method, which simultaneously takes into account all of the concerned variables, allows the separation of patients into subgroups, which may help clinicians make their diagnoses.