Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Segmentation”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 919 records · Page 51Linked to original sources
PubMed2003

Semiautomated segmentation of blood vessels using ellipse-overlap criteria: method and comparison to manual editing.

Two-dimensional intensity-based methods for the segmentation of blood vessels from computed-tomography-angiography data often result in spurious segments that originate from other objects whose intensity distributions overlap with those of the vessels. When segmented images include spurious segments, additional methods are required to select segments that belong to the target vessels. We describe a method that allows experts to select vessel segments from sequences of segmented images with little effort. Our method uses ellipse-overlap criteria to differentiate between segments that belong to different objects and are separated in plane but are connected in the through-plane direction. To validate our method, we used it to extract vessel regions from volumes that were segmented via analysis of isolabel-contour maps, and showed that the difference between the results of our method and manually-edited results was within inter-expert variability. Although the total editing duration for our method, which included user-interaction and computer processing, exceeded that of manual editing, the extent of user interaction required for our method was about a fifth of that required for manual editing.

Aneurysm↗
PubMed2004

A new 2D segmentation method based on dynamic programming applied to computer aided detection in mammography.

Mass segmentation plays a crucial role in computer-aided diagnosis (CAD) systems for classification of suspicious regions as normal, benign, or malignant. In this article we present a robust and automated segmentation technique--based on dynamic programming--to segment mass lesions from surrounding tissue. In addition, we propose an efficient algorithm to guarantee resulting contours to be closed. The segmentation method based on dynamic programming was quantitatively compared with two other automated segmentation methods (region growing and the discrete contour model) on a dataset of 1210 masses. For each mass an overlap criterion was calculated to determine the similarity with manual segmentation. The mean overlap percentage for dynamic programming was 0.69, for the other two methods 0.60 and 0.59, respectively. The difference in overlap percentage was statistically significant. To study the influence of the segmentation method on the performance of a CAD system two additional experiments were carried out. The first experiment studied the detection performance of the CAD system for the different segmentation methods. Free-response receiver operating characteristics analysis showed that the detection performance was nearly identical for the three segmentation methods. In the second experiment the ability of the classifier to discriminate between malignant and benign lesions was studied. For region based evaluation the area Az under the receiver operating characteristics curve was 0.74 for dynamic programming, 0.72 for the discrete contour model, and 0.67 for region growing. The difference in Az values obtained by the dynamic programming method and region growing was statistically significant. The differences between other methods were not significant.

Aged↗
PubMed2000

Knowledge-based segmentation of thoracic computed tomography images for assessment of split lung function.

The assessment of differential left and right lung function is important for patients under consideration for lung resection procedures such as single lung transplantation. We developed an automated, knowledge-based segmentation algorithm for purposes of deriving functional information from dynamic computed tomography (CT) image data. Median lung attenuation (HU) and area measurements were automatically calculated for each lung from thoracic CT images acquired during a forced expiratory maneuver as indicators of the amount and rate of airflow. The accuracy of these derived measures from fully automated segmentation was validated against those from segmentation using manual editing by an expert observer. A total of 1313 axial images were analyzed from 49 patients. The images were segmented using our knowledge-based system that identifies the chest wall, mediastinum, trachea, large airways and lung parenchyma on CT images. The key components of the system are an anatomical model, an inference engine and image processing routines, and segmentation involves matching objects extracted from the image to anatomical objects described in the model. The segmentation results from all images were inspected by the expert observer. Manual editing was required to correct 183 (13.94%) of the images, and the sensitivity, specificity, and accuracy of the knowledge-based segmentation were greater than 98.55% in classifying pixels as lung or nonlung. There was no significant difference between median lung attenuation or area values from automated and edited segmentations (p > 0.70). Using the knowledge-based segmentation method we can automatically derive indirect quantitative measures of single lung function that cannot be obtained using conventional pulmonary function tests.

Algorithms↗
PubMed1991

Site-specific recombinase genes in three Shigella subgroups and nucleotide sequences of a pinB gene and an invertible B segment from Shigella boydii.

Inversional switching systems in procaryotes are composed of an invertible DNA segment and a site-specific recombinase gene adjacent to or contained in the segment. Four related but functionally distinct systems have previously been characterized in detail: the Salmonella typhimurium H segment-hin gene (H-hin), phage Mu G-gin, phage P1 C-cin, and Escherichia coli e14 P-pin. In this article we report the isolation and characterization of three new recombinase genes: pinB, pinD, and defective pinF from Shigella boydii, Shigella dysenteriae, and Shigella flexneri, respectively. The genes pinB and pinD were detected by the complementation of a hin mutation of Salmonella and were able to mediate inversion of the H, P, and C segments. pinB mediated H inversion as efficiently as the hin gene did and mediated C inversion with a frequency three orders of magnitude lower than that of the cin gene. pinD mediated inversion of H and P segments with frequencies ten times as high as those for the genes intrinsic to each segment and mediated C inversion with a frequency ten times lower than that for cin. Therefore, the pinB and pinD genes were inferred to be different from each other. The invertible B segment-pinB gene cloned from S. boydii is highly homologous to the G-gin in size, organization, and nucleotide sequence of open reading frames, but the 5' constant region outside the segment is quite different in size and predicted amino acid sequence. The B segment underwent inversion in the presence of hin, pin, or cin. The defective pinF gene is suggested to hae the same origin as P-pin on e14 by the restriction map of the fragment cloned from a Pin+ transductant that was obtained in transduction from S. flexneri to E. coli delta pin.

Amino Acid Sequence↗
PubMed1993

Multiple double-stranded RNA segments are associated with virus particles infecting Trichomonas vaginalis.

Previous studies demonstrated that some isolates of the sexually transmitted protozoan Trichomonas vaginalis are infected with a nonsegmented, double-stranded RNA (dsRNA) virus. A reexamination of the total dsRNA extracted from several virus-harboring isolates indicated the presence of at least three dsRNAs with sizes ranging from 4.8 to 4.3 kbp. The double-stranded nature of each of the three segments was determined by hybridization experiments using riboprobes of opposite polarities obtained from cDNA generated to each of the segments. All three segments were present in agar clones originating from single organisms of T. vaginalis isolates, suggesting that the three segments were not the result of a mixed population of trichomonads harboring different sizes of dsRNA. The three segments were associated with CsCl-purified virus particles, as evidenced by electron microscopy, and RNAse treatment of the preparation containing virus particles did not destroy the dsRNAs. Finally, the individual dsRNA segments were purified for use as probes to determine whether the three dsRNAs shared any sequence homology. Each end-labeled dsRNA segment did not cross-hybridize to any of the other two segments, a finding consistent with the hybridization of labeled cDNAs to only the segments from which they were derived. These results show that the coding capacity of the dsRNA virus may be at least three times greater than that estimated earlier and illustrates further the complexity of this virus-parasite interrelationship.

Animals↗
PubMed2006

Hierarchy among viral RNA (vRNA) segments in their role in vRNA incorporation into influenza A virions.

The genome of influenza A viruses comprises eight negative-strand RNA segments. Although all eight segments must be present in cells for efficient viral replication, the mechanism(s) by which these viral RNA (vRNA) segments are incorporated into virions is not fully understood. We recently found that sequences at both ends of the coding regions of the HA, NA, and NS vRNA segments of A/WSN/33 play important roles in the incorporation of these vRNAs into virions. In order to similarly identify the regions of the PB2, PB1, and PA vRNAs of this strain that are critical for their incorporation, we generated a series of mutant vRNAs that possessed the green fluorescent protein gene flanked by portions of the coding and noncoding regions of the respective segments. For all three polymerase segments, deletions at the ends of their coding regions decreased their virion incorporation efficiencies. More importantly, these regions not only affected the incorporation of the segment in which they reside, but were also important for the incorporation of other segments. This effect was most prominent with the PB2 vRNA. These findings suggest a hierarchy among vRNA segments for virion incorporation and may imply intersegment association of vRNAs during virus assembly.

Animals↗
PubMedDate not supplied

Medical image segmentation with knowledge-guided robust active contours.

Medical image segmentation techniques typically require some form of expert human supervision to provide accurate and consistent identification of anatomic structures of interest. A novel segmentation technique was developed that combines a knowledge-based segmentation system with a sophisticated active contour model. This approach exploits the guidance of a higher-level process to robustly perform the segmentation of various anatomic structures. The user need not provide initial contour placement, and the high-level process carries out the required parameter optimization automatically. Knowledge about the anatomic structures to be segmented is defined statistically in terms of probability density functions of parameters such as location, size, and image intensity (eg, computed tomographic [CT] attenuation value). Preliminary results suggest that the performance of the algorithm at chest and abdominal CT is comparable to that of more traditional segmentation techniques like region growing and morphologic operators. In some cases, the active contour-based technique may outperform standard segmentation methods due to its capacity to fully enforce the available a priori knowledge concerning the anatomic structure of interest. The active contour algorithm is particularly suitable for integration with high-level image understanding frameworks, providing a robust and easily controlled low-level segmentation tool. Further study is required to determine whether the proposed algorithm is indeed capable of providing consistently superior segmentation.

Algorithms↗
PubMed2000

Anatomy of the right anterosuperior area (segment 8) of the liver: evaluation with helical CT during arterial portography.

PURPOSE: To evaluate the segmental anatomy of the right anterosuperior area (segment 8) of the liver by using helical computed tomography during arterial portography (CTAP). MATERIALS AND METHODS: Twenty-seven patients without lesions at segment 8 underwent helical CTAP. Three-dimensional portograms were reconstructed to verify the course of the portal veins. The number of subsegmental branches, in addition to the branching point and the distribution in segment 8, was assessed. RESULTS: In 25 (93%) patients, the dorsal branch of segment 8 gave rise to dorsally directed branches posterior to the right hepatic vein. In only four (25%) of 16 patients in whom the medial branch of segment 8 arose near the porta hepatis, the long paracaval portal branch of the caudate lobe extended upward above the interval between the middle and right hepatic veins. CONCLUSION: In most of the patients, the dorsal branches of segment 8 supplied the dorsocranial area of the right lobe posterior to the right hepatic vein. The paracaval portion of the caudate lobe was limited to below the interval between the middle and right hepatic veins in the majority of patients who showed medial branches of segment 8 arising near the porta hepatis. Recognition of this vascular anatomy is clinically important for preoperative evaluation of hepatic tumors in segment 8 because it may contribute to a safer surgical approach.

Adult↗
PubMed2007

Mechanisms underlying nutrient-induced segmentation in isolated guinea pig small intestine.

Mechanisms underlying nutrient-induced segmentation within the gut are not well understood. We have shown that decanoic acid and some amino acids induce neurally dependent segmentation in guinea pig small intestine in vitro. This study examined the neural mechanisms underlying segmentation in the circular muscle and whether the timing of segmentation contractions also depends on slow waves. Decanoic acid (1 mM) was infused into the lumen of guinea pig duodenum and jejunum. Video imaging was used to monitor intestinal diameter as a function of both longitudinal position and time. Circular muscle electrical activity was recorded by using suction electrodes. Recordings from sites of segmenting contractions showed they are always associated with excitatory junction potentials leading to action potentials. Recordings from sites oral and anal to segmenting contractions revealed inhibitory junction potentials that were time locked to those contractions. Slow waves were never observed underlying segmenting contractions. In paralyzed preparations, intracellular recording revealed that slow-wave frequency was highly consistent at 19.5 (SD 1.4) cycles per minute (c/min) in duodenum and 16.6 (SD 1.1) c/min in jejunum. By contrast, the frequencies of segmenting contractions varied widely (duodenum: 3.6-28.8 c/min, median 10.8 c/min; jejunum: 3.0-27.0 c/min, median 7.8 c/min) and sometimes exceeded slow-wave frequencies for that region. Thus nutrient-induced segmentation contractions in guinea pig small intestine do not depend on slow-wave activity. Rather they result from a neural circuit producing rhythmic localized activity in excitatory motor neurons, while simultaneously activating surrounding inhibitory motor neurons.

Action Potentials↗
PubMed1985

Characterization of segmental collateral blood flow in the small intestine.

Collateral blood flow and the pressures and resistances determining that flow were measured between two adjacent segments of canine jejunum following acute occlusion of the arterial branch perfusing one of the segments (the "recipient" segment). Collateral flow was approximately 55% of control flow in the recipient segment. This flow was provided by an equal increment in arterial flow to the nonischemic ("donor") segment, such that pressures, resistances, and flows in the donor segment were not affected. Virtually all of the total collateral flow was derived from precapillary channels and was therefore available to the capillary bed of the recipient segment. Collateral flow was adequate to maintain the recipient segment in a nonischemic state, as indicated by the absence of a reactive hyperemia following release of the arterial occlusion. Selective occlusions of intramural or extramural collateral channels indicate that about two-thirds of the total collateral flow is derived from the extramural (marginal) vessels, while the remainder is supplied by intramural collaterals. For the most part collateral flow between adjacent segments is determined simply by the pressure gradient between connecting collateral channels.

Animals↗
PubMed2006

Progressive epicardial coronary blood flow reduction fails to produce ST-segment depression at normal heart rates.

ST-segment depression is commonly seen in patients with acute coronary syndromes. Most authors have attributed it to transient reductions in coronary blood flow due to nonocclusive thrombus formation on a disrupted atherosclerotic plaque and dynamic focal vasospasm at the site of coronary artery stenosis. However, ST-segment depression was never reproduced in classic animal models of coronary stenosis without the presence of tachycardia. We hypothesized that ST-segment depression occurring during acute coronary syndromes is not entirely explained by changes in epicardial coronary artery resistance and thus evaluated the effect of a slow, progressive epicardial coronary artery occlusion on the ECG and regional myocardial blood flow in anesthetized pigs. Slow, progressive occlusion over 72 min (SD 27) of the left anterior descending coronary artery in 20 anesthetized pigs led to a 90% decrease in coronary blood flow and the development of ST-segment elevation associated with homogeneous and transmural myocardial blood flow reductions, confirmed by microspheres and myocardial contrast echocardiography. ST-segment depression was not observed in any ECG lead before the development of ST-segment elevation. At normal heart rates, progressive epicardial stenosis of a coronary artery results in myocardial ischemia associated with homogeneous, transmural reduction in regional myocardial blood flow and ST-segment elevation, without preceding ST-segment depression. Thus, in coronary syndromes with ST-segment depression and predominant subendocardial ischemia, factors other than mere increases in epicardial coronary resistance must be invoked to explain the heterogeneous parietal distribution of flow and associated ECG changes.

Angina, Unstable↗
PubMed1979

Auxotonic contractions in cardiac muscle segments.

The dynamics of segment shortening have been measured in the central regions of isolated papillary muscles during muscle isometric and after-loaded isotonic contractions. Segment lengths are inferred from muscle cross-sectional area using an assumption that the segments remain isovolumic. Area is assessed with a magnetic induction technique. Infused microspheres have been used as visual markers to corroborate the segment length measurement. The results confirm the existence of major segmental shortening during muscle isometric conditions. However, the time course of shortening is not the same as that of force development. Rather, the segments remain shortened until after force has fallen significantly from its peak value. This behavior appears in the force-segment length plane as counterclockwise loops. The relationship of peak force to segment length has been determined and found to depend on the mechanical conditions under which the muscle is equilibrated. These results demonstrate the utility of the new technique and indicate central segment behavior that is substantially different from that observed for the whole muscle.

Animals↗
PubMed1983

Comparison of arterial wall mechanics using ring and cylindrical segments.

Thin rings and intact cylindrical segments of canine carotid and iliac arteries were used to determine wall mechanics. Measurements of force and length were obtained from the ring segments, whereas measurements of pressure and diameter were obtained from the cylindrical segments under conditions of active (147 mM K+) and passive smooth muscle (Ca2+ free and 2 mM ethyleneglycolbis (beta-aminoethylether)-N,N'-tetraacetic acid). These measurements were normalized to values of segment stress and strain. Under passive conditions stress-strain relations for the rings appeared to be stiffer than those obtained using cylindrical segments. Pressure-diameter curves computed using force-length data from the rings were shifted to higher values of diameter compared with values from the intact segments at all pressure levels. Passive mechanics derived from measurements on ring segments yielded poor estimates of mechanics derived from intact segments. Despite this finding, values of active force development from the two sample geometries were similar. No statistically significant differences were found in values of maximum force development expressed in terms of sample cross-sectional area. Some differences in values of active force development at low values of muscle length were found. The latter were probably related to the differences in passive mechanics and the procedure used to normalize muscle length. Reasonable values of active force development can be obtained from ring segments.

Animals↗
PubMed1992

T-Q, S-T segment mapping and hyperemia in reperfused pig heart with ischemic preconditioning.

The effects of ischemic preconditioning on epicardial T-Q and S-T segment mapping, local activation, and coronary blood flow were analyzed in nine barbiturate-anesthetized pigs during four coronary occlusion (5 min)-reperfusion (20 min) sequences. In seven sham pigs, one occlusion was performed after a control period of 75 min. The first reperfusion induced a marked coronary hyperemia [11 +/- 4 ml/min (baseline) to 33 +/- 16 ml/min, P < 0.005] and a rapid recovery (30 to 150 s) of epicardial activation delays, T-Q segment depression, and S-T segment elevation in the ischemia area. This recovery was transiently associated with enlargement of intersite T-Q potential variability (alpha: 2.5 +/- 0.6 to 3.4 +/- 0.7 mV, P < 0.05), T-Q segment overshoot to +1.4 +/- 0.9 mV, and S-T segment reelevation. A brief T-Q segment depression (-2.3 +/- 0.9 mV) occurred during early reperfusion in 60 of 91 electrodes overlying the normal myocardium. Compared with the first, the fourth occlusion induced lower S-T segment elevation (3.4 +/- 2.0 to 1.7 +/- 1.9 mV, P < 0.05), and the fourth reperfusion elicited a faster reversal of T-Q segment dispersion (53 +/- 21 to 43 +/- 16 s, P < 0.05), S-T segment elevation (149 +/- 101 to 81 +/- 45 s, P < 0.05), and coronary hyperemia (8 +/- 2 to 5 +/- 1 min, P < 0.05). This trend of changes was not observed during a fourth occlusion in sham pigs.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed1998

Spinal segments communicating resting sympathetic activity to postganglionic nerves of the stellate ganglion.

It has been shown earlier using sympathetic reflexes and anatomic techniques that preganglionic neurons controlling different effectors occupy wide and overlapping ranges of adjacent segments in the spinal cord (cardiac: T1-T7, vertebral: T2-T8). Because, however, the majority of preganglionic neurons are silent at resting states, the present study was designed to estimate the segmental map of subsets of these neurons including only those active at rest using simultaneous recordings from the inferior cardiac and vertebral nerves, under chloralose-urethan or urethan anesthesia. In 22 cats, thoracic white rami T1-T8 were cut in a sequential manner. Three-minute-long data segments were recorded between sectionings and analyzed in the frequency domain using the fast Fourier transform. We found that cardiac and vertebral active maps involved segments T3-T5 and T4-T8, respectively. In individual experiments, however, most of the power of rhythmic activity originated from only one or two segments and the dominant segments for the two nerves never overlapped. Moreover, the separation between dominant segments generating cardiac and vertebral nerve discharges was wider and the distribution of tonically active preganglionic neurons projecting to each nerve was narrower under urethan than chloralose-urethan anesthesia. We conclude that the proportion of active to quiescent preganglionic neurons regulating cardiac and vertebral nerve discharges varies from spinal segment to segment and that active neurons projecting to these nerves are nonoverlapping.

Animals↗
PubMed1996

Segmental body composition assessed by bioelectrical impedance analysis and DEXA in humans.

The present study assessed the relative contribution of each body segment to whole body fat-free mass (FFM) and impedance and explored the use of segmental bioelectrical impedance analysis to estimate segmental tissue composition. Multiple frequencies of whole body and segmental impedances were measured in 51 normal and overweight women. Segmental tissue composition was independently assessed by dual-energy X-ray absorptiometry. The sum of the segmental impedance values corresponded to the whole body value (100.5 +/- 1.9% at 50 kHz). The arms and legs contributed to 47.6 and 43.0%, respectively, of whole body impedance at 50 kHz, whereas they represented only 10.6 and 34.8% of total FFM, as determined by dual-energy X-ray absorptiometry. The trunk averaged 10.0% of total impedance but represented 48.2% of FFM. For each segment, there was an excellent correlation between the specific impedance index (length2/impedance) and FFM (r = 0.55, 0.62, and 0.64 for arm, trunk, and leg, respectively). The specific resistivity was in a similar range for the limbs (159 +/- 23 cm for the arm and 193 +/- 39 cm for the leg at 50 kHz) but was higher for the trunk (457 +/- 71 cm). This study shows the potential interest of segmental body composition by bioelectrical impedance analysis and provides specific segmental body composition equations for use in normal and overweight women.

Absorptiometry, Photon↗
PubMedDate not supplied

Differential diagnosis of aging, dementia of the Alzheimer type and depression with EEG-segmentation.

EEG segmentation can be used to measure altered brain function in aging and diseases of the brain. The parameter 'number of different segments' makes clear how many different potential fields are involved in brain activity during a given period of time. It should represent effects of aging and disease. To prove this assumption, 11 young and 10 aged controls, 12 patients with mild dementia of the Alzheimer type (DAT), 10 young and 12 aged patients with endogenous depression were included in the study. The number of different segments in the beta frequency band between 16 and 19.75 Hz was measured according to the theory of Lehmann et al. [Clinical Neurophysiology 1987;67:271-288], and the segments were classified by their location on the scalp. The Mann-Whitney U test was used for statistical comparison. Aged controls had more different segments than young controls (n = 21, U = 14, p < 0.0038). Patients with DAT had less different segments than healthy aged controls (n = 22, U = 18.5, p < 0.0061). Aged patients with endogenous depression had more different segments than patients with mild DAT (n = 24, U = 32, p < 0.021). The reduction of the number of different segments in DAT compared to controls and patients suffering from depression may be helpful for differential diagnosis. The higher number of different segments in aged versus young controls could be interpreted as a sign of increased complexity in the aged brain.

Adult↗
PubMed2006

Anatomical study and proportional analysis of the pig spleen arterial segments.

A systematic study was performed in 31 pig spleens to provide an analysis of the splenic arterial segmentation and an analysis of the proportional area of each segment. There were two independent arterial segments in all cases, a dorsal and a ventral. The dorsal segment was smaller than the ventral segment, presenting 16.81 and 83.19% as the median percentage of the total area of the spleen, respectively. In conclusion, the pig arterial segments are quite similar to those of dog, buffalo, goat, and sheep, where there is a small dorsal segment and a large ventral segment, and somewhat similar to the human spleen, where there are two segments as well, though with a similar area.

Animals↗