PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Automated image segmentation”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11Linked to original sources
PubMed1996

Tumour volume determination from MR images by morphological segmentation.

Accurate tumour volume measurement from MR images requires some form of objective image segmentation, and therefore a certain degree of automation. Manual methods of separating data according to the various tissue types which they are thought to represent are inherently prone to operator subjectivity and can be very time consuming. A segmentation procedure based on morphological edge detection and region growing has been implemented and tested on a phantom of known adjustable volume. Comparisons have been made with a traditional data thresholding procedure for the determination of tumour volumes on a set of patients with intracerebral glioma. The two methods are shown to give similar results, with the morphological segmentation procedure having the advantages of being automated and faster.

Brain Neoplasms↗
PubMed1988

Medical images and automated interpretation.

The application of automated interpretation to medical images is discussed and the main methods of medical imaging are briefly described. The factors behind the process of human clinical interpretation are also considered. Because human interpretation can be aided by processing of the raw image, the standard methods of image processing are mentioned. Automated interpretation of images in other fields is relevant, and segmentation of images is an important initial part of the process. There are already some applications of automated interpretation of medical images. Some are more complete than others. This will undoubtedly be an important developing field of work which will draw on experience from other areas, and be spurred on by the increasing complexity of medical imaging methods and shortage of expertise for human interpretation.

Artificial Intelligence↗
PubMed2006

Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification.

We present a method for automated segmentation of the vasculature in retinal images. The method produces segmentations by classifying each image pixel as vessel or nonvessel, based on the pixel's feature vector. Feature vectors are composed of the pixel's intensity and two-dimensional Gabor wavelet transform responses taken at multiple scales. The Gabor wavelet is capable of tuning to specific frequencies, thus allowing noise filtering and vessel enhancement in a single step. We use a Bayesian classifier with class-conditional probability density functions (likelihoods) described as Gaussian mixtures, yielding a fast classification, while being able to model complex decision surfaces. The probability distributions are estimated based on a training set of labeled pixels obtained from manual segmentations. The method's performance is evaluated on publicly available DRIVE (Staal et al., 2004) and STARE (Hoover et al., 2000) databases of manually labeled images. On the DRIVE database, it achieves an area under the receiver operating characteristic curve of 0.9614, being slightly superior than that presented by state-of-the-art approaches. We are making our implementation available as open source MATLAB scripts for researchers interested in implementation details, evaluation, or development of methods.

Algorithms↗
PubMed1998

Automated lung segmentation in digital lateral chest radiographs.

We are developing a fully automated computerized scheme for segmenting the lung fields in digital lateral chest radiographs. Existing computer-aided diagnostic (CAD) schemes and automated lung segmentation methods have focused exclusively on the posteroanterior view, despite the diagnostic importance of the lateral view. Information from the lateral radiograph is routinely incorporated by radiologists in their decision-making process, and thus computer analysis of lateral images may potentially add another dimension to current CAD schemes. Automated analysis of the lung fields in lateral images will necessarily require accurate segmentation. Our scheme employs an initial procedure to eliminate external and subcutaneous pixels. Global gray-level histogram analysis then allows for the identification of a range of gray-level thresholds. An iterative gray-level thresholding scheme is implemented using this range of thresholds to construct a series of binary images in which contiguous regions are identified and geometrically analyzed. Regions determined to be outside the lungs are prevented from contributing to binary images at later iterations. Adaptive local gray-level thresholding is applied along the initial contour that results from the global thresholding procedure to extend the contour closer to the true lung borders. This local thresholding method uses regions of interest of various dimensions, depending on the enclosed anatomy. Smoothing and polynomial curve fitting complete the segmentation. A database of 100 normal and 100 abnormal lateral images was analyzed. Quantitative comparison of computer-segmented lung regions with lung regions manually delineated by two radiologists indicated that 83% and 84% of normal and abnormal images, respectively, displayed segmentation contours within three standard deviations of the mean inter-radiologist contour degree-of-overlap value.

Automation↗
PubMed2005

Cortical gray and white brain tissue volume in adolescents and adults with autism.

BACKGROUND: A number of studies have found brain enlargement in autism, but there is disagreement as to whether this enlargement is limited to early development or continues into adulthood. In this study, cortical gray and white tissue volumes were examined in a sample of adolescents and adults with autism who had demonstrated total brain enlargement in a previous magnetic resonance imaging (MRI) study. METHODS: An automated tissue segmentation program was applied to structural MRI scans to obtain volumes of gray, white, and cerebrospinal fluid (CSF) tissue on a sample of adolescent and adult males ages 13-29 with autism (n = 23) and controls (n = 15). Regional differences for brain lobes and brain hemispheres were also examined. RESULTS: Significant enlargement in gray matter volume was found for the individuals with autism, with a disproportionate increase in left-sided gray matter volume. Lobe volume enlargements were detected for frontal and temporal, but not parietal or occipital lobes, in the subjects with autism. Age and nonverbal IQ effects on tissue volume were also observed. CONCLUSIONS: These findings give evidence for left-lateralized gray tissue enlargement in adolescents and adults with autism, and demonstrate a regional pattern of cortical lobe volumes underlying this effect.

Adolescent↗
PubMed1977

Scene segmentation techniques for the analysis of routine bone marrow smears from acute lymphoblastic leukemia patients.

Automated analysis of lymphoblast cell morphology is being evaluated as a basis for predicting the response to therapy of patients with acute lymphoblastic leukemia. A new technique of scene segmentation particularly applicable to the "cluttered" images of cells in routine bone marrow smears is described. Morphologic characteristics of lymphoblasts found in bone marrow smears made at time of diagnosis were measured by an automated, interactive image-processing system using the new scene segmentation technique. These characteristics, on a patient by patient basis, are being compared to remission length and survival data to develop and test new prognostic methods.

Autoanalysis↗
PubMed2000

The relation between tendency for psychopathology and reduced frontal brain volume in healthy people.

OBJECTIVE: We hypothesized that tendency toward psychopathology is associated with lower frontotemporal volumes. BACKGROUND: Although there is considerable evidence for structural abnormalities in patients with major psychiatric disorders and increased recognition that neural substrates may underlie individual differences in personality, there have been no studies in healthy people attempting to relate personality to volumetric measures of brain structure. METHOD: We used magnetic resonance imaging with an advanced method for automated segmentation of cranial compartments to gray matter, white matter, and cerebrospinal fluid. We examined the relation between frontal and temporal lobe volumes and Minnesota Multiphasic Personality Inventory measures of tendency toward psychopathology in 59 healthy individuals. RESULTS: As hypothesized, higher scores on the clinical scales were associated with lower average frontal lobe volume. When the sample was divided according to sex, however, these correlations were significant in men (n = 29) but not in women (n = 30). The highest correlation was observed between lower frontal white matter volume in men and high schizophrenia scale score (r[27] = -0.59, p <0.001). CONCLUSIONS: The findings suggest that personality dimensions in healthy people can be linked to neural substrates, which can potentially serve as endophenotypic markers of disposition to psychopathology. The sexually dimorphic effects are consistent with gender-related differences in the clinical manifestations of psychiatric disorders and may suggest sex hormone modulation of the psychopathologic processes.

Adult↗
PubMed2001

Segmenting brain white matter, gray matter and cerebro-spinal fluid using diffusion tensor-MRI derived indices.

Based on its ability to provide quantitative information about tissue microstructure, diffusion tensor magnetic resonance imaging (DT-MRI) might be a valuable approach to improve the reliability of segmentation of the various brain tissues. In this study, a fully automated and easy-to-implement technique based on 2D histogram analysis of DT-MRI derived images was used to segment white and gray matter of the brain from 10 healthy subjects (aged = 27-56 years). The results obtained with this novel segmentation strategy were compared to those achieved by two experienced observers using an operator-dependent segmentation on the dual-echo scans. Visual inspection of the segmented tissues from a third senior observer disclosed that the automated technique worked properly on all images from all subjects and was more accurate than the human raters in defining thalamus white and gray matter portions as well as in tissue classification at the external brain edge. In addition, this segmentation technique resulted in an average gray/white matter ratio similar to that reported by post-mortem assessment. The application of the operator-dependent segmentation strategy was extremely time-consuming and the two observers achieved poorly reproducible results. Segmenting brain white and gray matter using information from DT-MRI proved to be an accurate approach with the potential for improving the understanding of the pathophysiology of many neurologic conditions.

Anisotropy↗
PubMed2002

Quantitative analysis of MRI signal abnormalities of brain white matter with high reproducibility and accuracy.

PURPOSE: To assess the reproducibility and accuracy compared to radiologists of three automated segmentation pipelines for quantitative magnetic resonance imaging (MRI) measurement of brain white matter signal abnormalities (WMSA). MATERIALS AND METHODS: WMSA segmentation was performed on pairs of whole brain scans from 20 patients with multiple sclerosis (MS) and 10 older subjects who were positioned and imaged twice within 30 minutes. Radiologist outlines of WMSA on 20 sections from 16 patients were compared with the corresponding results of each segmentation method. RESULTS: The segmentation method combining expectation-maximization (EM) tissue segmentation, template-driven segmentation (TDS), and partial volume effect correction (PVEC) demonstrated the highest accuracy (the absolute value of the Z-score was 0.99 for both groups of subjects), as well as high interscan reproducibility (repeatability coefficient was 0.68 mL in MS patients and 1.49 mL in aging subjects). CONCLUSION: The addition of TDS to the EM segmentation and PVEC algorithms significantly improved the accuracy of WMSA volume measurements, while also improving measurement reproducibility.

Aging↗
PubMed1987

Suppression of neurofilament degradation by protease inhibitors in experimental spinal cord injury.

Intraperitoneal administration of the neutral protease inhibitors leupeptin and E-64c substantially suppressed the degradation of neurofilament proteins (NFP) at the site of mechanical insult and secondary axonal degeneration, and facilitated the recovery of motor functions in acute spinal cord injury in rats. The drug effects were assessed by sodium dodecyl sulphate polyacrylamide gel electrophoresis of NFP fractions from the injured tissue and by morphometry of degenerating axons revealed by the Fink-Heimer method in distal spinal cord segments with the aid of an automated image analyzer. The role of calcium-activated neutral proteases in acute central nervous tissue damage and potential use of protease inhibitors as therapeutic modalities are discussed.

Animals↗
PubMed2005

Automated segmentation of neuroanatomical structures in multispectral MR microscopy of the mouse brain.

We present the automated segmentation of magnetic resonance microscopy (MRM) images of the C57BL/6J mouse brain into 21 neuroanatomical structures, including the ventricular system, corpus callosum, hippocampus, caudate putamen, inferior colliculus, internal capsule, globus pallidus, and substantia nigra. The segmentation algorithm operates on multispectral, three-dimensional (3D) MR data acquired at 90-microm isotropic resolution. Probabilistic information used in the segmentation is extracted from training datasets of T2-weighted, proton density-weighted, and diffusion-weighted acquisitions. Spatial information is employed in the form of prior probabilities of occurrence of a structure at a location (location priors) and the pairwise probabilities between structures (contextual priors). Validation using standard morphometry indices shows good consistency between automatically segmented and manually traced data. Results achieved in the mouse brain are comparable with those achieved in human brain studies using similar techniques. The segmentation algorithm shows excellent potential for routine morphological phenotyping of mouse models.

Algorithms↗
PubMed2003

Automatic brain tumor segmentation by subject specific modification of atlas priors.

RATIONALE AND OBJECTIVES: Manual segmentation of brain tumors from magnetic resonance images is a challenging and time-consuming task. An automated system has been developed for brain tumor segmentation that will provide objective, reproducible segmentations that are close to the manual results. Additionally, the method segments white matter, grey matter, cerebrospinal fluid, and edema. The segmentation of pathology and healthy structures is crucial for surgical planning and intervention. MATERIALS AND METHODS: The method performs the segmentation of a registered set of magnetic resonance images using an expectation-maximization scheme. The segmentation is guided by a spatial probabilistic atlas that contains expert prior knowledge about brain structures. This atlas is modified with the subject-specific brain tumor prior that is computed based on contrast enhancement. RESULTS: Five cases with different types of tumors are selected for evaluation. The results obtained from the automatic segmentation program are compared with results from manual and semi-automated methods. The automated method yields results that have surface distances at roughly 1-4 mm compared with the manual results. CONCLUSION: The automated method can be applied to different types of tumors. Although its performance is below that of the semi-automated method, it has the advantage of requiring no user supervision.

Automation↗
PubMed2002

3-D active appearance models: segmentation of cardiac MR and ultrasound images.

A model-based method for three-dimensional image segmentation was developed and its performance assessed in segmentation of volumetric cardiac magnetic resonance (MR) images and echocardiographic temporal image sequences. Comprehensive design of a three-dimensional (3-D) active appearance model (AAM) is reported for the first time as an involved extension of the AAM framework introduced by Cootes et al. The model's behavior is learned from manually traced segmentation examples during an automated training stage. Information about shape and image appearance of the cardiac structures is contained in a single model. This ensures a spatially and/or temporally consistent segmentation of three-dimensional cardiac images. The clinical potential of the 3-D AAM is demonstrated in short-axis cardiac MR images and four-chamber echocardiographic sequences. The method's performance was assessed by comparison with manually identified independent standards in 56 clinical MR and 64 clinical echo image sequences. The AAM method showed good agreement with the independent standard using quantitative indexes of border positioning errors, endo- and epicardial volumes, and left ventricular mass. In MR, the endocardial volumes, epicardial volumes, and left ventricular wall mass correlation coefficients between manual and AAM were R2 = 0.94, 0.97, 0.82, respectively. For echocardiographic analysis, the area correlation was R2 = 0.79. The AAM method shows high promise for successful application to MR and echocardiographic image analysis in a clinical setting.

Echocardiography, Three-Dimensional↗
PubMed1997

ECG-gated three-dimensional intravascular ultrasound: feasibility and reproducibility of the automated analysis of coronary lumen and atherosclerotic plaque dimensions in humans.

BACKGROUND: Automated systems for the quantitative analysis of three-dimensional (3D) sets of intravascular ultrasound (IVUS) images have been developed to reduce the time required to perform volumetric analyses; however, 3D image reconstruction by these nongated systems is frequently hampered by cyclic artifacts. METHODS AND RESULTS: We used an ECG-gated 3D IVUS image acquisition workstation and a dedicated pullback device in atherosclerotic coronary segments of 30 patients to evaluate (1) the feasibility of this approach of image acquisition, (2) the reproducibility of an automated contour detection algorithm in measuring lumen, external elastic membrane, and plaque+media cross-sectional areas (CSAs) and volumes and the cross-sectional and volumetric plaque+media burden, and (3) the agreement between the automated area measurements and the results of manual tracing. The gated image acquisition took 3.9+/-1.5 minutes. The length of the segments analyzed was 9.6 to 40.0 mm, with 2.3+/-1.5 side branches per segment. The minimum lumen CSA measured 6.4+/-1.7 mm2, and the maximum and average CSA plaque+media burden measured 60.5+/-10.2% and 46.5+/-9.9%, respectively. The automated contour-detection required 34.3+/-7.3 minutes per segment. The differences between these measurements and manual tracing did not exceed 1.6% (SD<6.8%). Intraobserver and interobserver differences in area measurements (n=3421; r=.97 to.99) were <1.6% (SD<7.2%); intraobserver and interobserver differences in volumetric measurements (n=30; r=.99) were <0.4% (SD<3.2%). CONCLUSIONS: ECG-gated acquisition of 3D IVUS image sets is feasible and permits the application of automated contour detection to provide reproducible measurements of the lumen and atherosclerotic plaque CSA and volume in a relatively short analysis time.

Adult↗
PubMed1996

Automated measurement of infarct size with scanned images of triphenyltetrazolium chloride-stained rat brains.

BACKGROUND AND PURPOSE: The extent of brain infarction after local cerebral ischemia is frequently assessed with the mitochondrial activity indicator 2,3,5-triphenyltetrazolium chloride (TTC). We describe an automated procedure for analysis of infarct size in TTC-stained rat brains. METHODS: Rats were subjected to middle cerebral artery occlusion and killed after 24 to 36 hours, and their brains were processed for TTC staining. Digital images of coronal sections from these brains (n > 50) were acquired with a desktop color scanner. The resulting images were divided into red, blue, and green component images. Total brain and infarct areas were automatically determined on the basis of total pixel intensity and area after segmentation of the red and green images, respectively. Automated measurements were compared with those made with a video camera-based image acquisition system that required manual tracing of lesion boundaries. RESULTS: The spatial resolution of scanned brain images (approximately equal to 200 microns) was comparable to that of the camera-based system and provided sufficient detail to recognize infarct boundaries and neuroanatomical features. Scanner-based acquisition and analysis were faster than with the camera-based method. The green component image accurately distinguished infarcted from normal brain, and the red component image represented total brain dimensions. Infarct measurements obtained by the automated method correlated closely with those from conventional apparatus (R2 = .89, P < .001). Intraobserver reliability with the automated method (R2 = 1.00) was higher than with the conventional method (R2 = .77). CONCLUSIONS: Infarct size after middle cerebral artery occlusion in the rat can be rapidly and reproducibly assessed with inexpensive scanning equipment and automated image analysis of TTC-stained brains.

Animals↗
PubMed1999

Fast, accurate, and reproducible automatic segmentation of the brain in T1-weighted volume MRI data.

A new fast automated algorithm has been developed to segment the brain from T1-weighted volume MR images. The algorithm uses automated thresholding and morphological operations. It is fully three-dimensional and therefore independent of scan orientation. The validity and the performance of the algorithm were evaluated by comparing the automatically calculated brain volume with semi-automated measurements in 10 subjects, by calculating the brain volume from repeated scans in another 10 subjects, and by visual inspection. The mean and standard deviation of the difference between semi-automated and automated measurements were 0.56% and 2.8% of the mean brain volume, respectively, which is within inter-observer variability of the semi-automated method. The mean and standard deviation of the difference between the total volumes calculated from repeated scans were 0.40% and 1.2% of the mean brain volume, respectively. Good results were also obtained from a scan of abnormal brains.

Adult↗
PubMed1993

Size of cortical bone and relationship to bone mineral density assessed by quantitative computed tomography image segmentation.

RATIONALE AND OBJECTIVES: The accuracy of the measurement of the size of cortical bone on computed tomography (CT) images of human vertebrae was evaluated using an automated contour detection and segmentation procedure. METHODS: Forty human lumbar vertebrae were scanned using 8-mm slices and an automated detection for definition of trabecular and cortical region of interest. The vertebrae were embedded in a polyester resin and 8-mm-thick midvertebral specimens were excised using a diamond circular saw. Contact radiographs of these specimens were performed and, after photograph magnification, the cortical area was measured using computerized planimetry. RESULTS: Cortical area measured on CT images was highly correlated with the area measured by planimetry on the specimens (r = .91; P < .001) with, however, a systematic over-estimation. A significant relationship was found between density and width of the cortex (r = .56; P < .001). CONCLUSIONS: Computed tomography is able to assess the size of cortical bone in human vertebrae, but a threshold detection algorithm, as used in the current study, is not adequate to obtain the precise anatomic dimensions.

Aged↗