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 1,621 records · Page 90Linked to original sources
PubMed2004

Segmental motion adjacent to an instrumented lumbar fusion: the effect of extension of fusion to the sacrum.

STUDY DESIGN: We present an in vitro biomechanical comparison of adjacent segment motion at the cranial segment (L3-L4) for an L4-L5 versus an L4-S1 fusion model using cadaveric lumbosacral spines. OBJECTIVES: The purpose is to determine the biomechanical effect on the unfused cranial segment of extending a short lumbar fusion to the sacrum versus stopping at L5. SUMMARY OF BACKGROUND DATA: Radiographic evidence of adjacent segment degeneration can occur as a late sequela in patients following lumbar and lumbosacral spinal fusions. It is believed that altered biomechanics adjacent to the fusion construct contribute to these degenerative changes. Little is known regarding changes in cranial adjacent segment mechanics resulting from inclusion of the sacrum compared to ending a fusion at L5. METHODS: Seven human cadaveric lumbosacral spines were instrumented with pedicle screws at L4, L5, and S1. Rods were placed from L4-L5 and from L4-S1 to simulate the corresponding fusion models. A material testing system was used to apply load-controlled moments to the spines in flexion-extension, lateral bending, and axial rotation. Electromagnetic sensors were used to record 6 df motion across the L3-L4, L4-L5, and L5-S1 motion segments. Angular displacements were recorded and system stiffness was calculated for each spine and construct. A paired sample t test was used to determine significance of recorded differences. RESULTS: Under flexion-extension loading, the angular displacement in the sagittal plane at L3-L4 for the L4-S1 model was 9.0 degrees compared to 7.8 degrees for the L4-L5 model (+15%; P = 0.002). Under lateral bending loading, L3-L4 motion in the coronal plane for the L4-S1 model was 12.8 degrees and was 14.5 degrees for the L4-L5 model (-12%; P = 0.002). In axial rotation testing, L3-L4 torsional motion for the L4-S1 model was equivalent to the L4-L5 model. Overall system stiffness increased for the L4-S1 model compared with the L4-L5 model. CONCLUSIONS: In this load-controlled model, extending fusion across L5-S1 did not consistently increase motion at L3-L4. While it may be difficult to translate this finding to a clinical setting, avoiding fusion to the sacrum in a lower lumbar fusion may not provide significant benefit from the standpoint of avoiding adjacent segment disease.

Adult↗
PubMed2005

Selective thoracic fusion with segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis: more than 5-year follow-up.

STUDY DESIGN: Retrospective study. OBJECTIVES: To evaluate outcomes of selective thoracic fusion with segmental pedicle screw fixation in thoracic idiopathic scoliosis with a minimum 5-year follow-up. SUMMARY OF BACKGROUND DATA: Segmental pedicle screw fixation has been proven to enable true segmental control and greater correction both in coronal and sagittal planes of scoliosis. However, there is no long-term study of selective thoracic fusion with segmental pedicle screw fixation in thoracic idiopathic scoliosis. METHODS: Two hundred and three thoracic idiopathic scoliosis patients with 236 thoracic curves subject to selective thoracic fusion with segmental pedicle screw fixation were analyzed. Mean patient age at the time of operation was 13.8 years (range: 8.9-18). RESULTS: The preoperative thoracic curve of 51 +/- 12 degrees was corrected to 16 +/- 7 degrees (69% correction, 3% loss of correction) at the most recent follow-up. The noninstrumented lumbar curve of 30 +/- 10 degrees was corrected to 10 +/- 8 degrees (66% correction, 5% loss of correction) at the most recent follow-up. The preoperative thoracic kyphosis of 18 +/- 11 degrees and the lumbar lordosis of 43 +/- 10 degrees were improved to 23 +/- 8 degrees and 46 +/- 9 degrees , respectively, at the most recent follow-up. There was no junctional kyphosis at the most recent follow-up. Coronal decompensation at the most recent follow-up occurred in 10 patients. Postoperative adding-on occurred in 17 patients, who were fused two levels short of the neutral vertebra. Of the 2,867 thoracic pedicle screws inserted in the thoracic level, 43 screws were found to be malpositioned (1.5%), but they did not cause neurologic complications or adversely affect the long-term results. CONCLUSIONS: Selective thoracic fusion with segmental pedicle screw fixation in thoracic idiopathic scoliosis had satisfactory radiographic and clinical outcomes after surgery and has been well-maintained for minimum 5-year follow-up. It is a safe and effective method for preservation of lumbar motion segments as well as for restoration and maintenance of both coronal and sagittal alignment.

Adolescent↗
PubMed2006

Mastication causing segmental spinal motion in common cervical orthoses.

STUDY DESIGN: In vivo fluoroscopic quantification of segmental cervical spinal motion in asymptomatic volunteers during mastication. OBJECTIVE: To quantify the degree of segmental cervical spine motion in patients during mastication and while wearing several commonly used cervical orthoses. SUMMARY OF BACKGROUND DATA: Cervical orthoses are routinely used to stabilize the cervical spine after trauma or cervical fusion and are, in some cases, prescribed to be worn at all times, including during eating. METHODS: Seven volunteers with an average age of 31 years (range, 26-42 years) had 5 seconds of continuous lateral cervical fluoroscopic imaging while they chewed gum without any cervical orthosis (control) and while wearing a soft collar turned forward then backward, a Philadelphia collar, a Miami J collar, or a two-poster brace. Still images were created from the fluoroscopy video, which allowed for quantification of the amount of segmental motion. RESULTS: All cervical collars produced statistically greater motion at occiput-C1 and C1-C2 when compared with no collar. The motion was greatest at the occiput-C1 junction and decreased to the C4 level. No motion was detected in any subject in any brace below C4. The two-poster brace produced the most segmental motion at occiput-C1 (6.3 degrees +/- 2.0 degrees) compared with the soft collar turned backwards (1.9 degrees +/- 1.9 degrees). No subject had any segmental motion below C2 in the soft collar. No segmental spinal motion was observed without a cervical brace. CONCLUSIONS: To limit upper cervical spine segmental motion during mastication, for patients with unstable cervical spines, we propose a two-poster brace with removal of the mandibular component to allow for free mandibular action. For stable spines, we propose a soft collar turned backwards. Patients should be educated with this information.

Adult↗
PubMed2006

The biomechanics of 1, 2, and 3 levels of vertebral augmentation with polymethylmethacrylate in multilevel spinal segments.

STUDY DESIGN: Experimental biomechanics of multilevel segments with 0, 1, 2, and 3 vertebral levels of polymethylmethacrylate augmentation. OBJECTIVE: To compare multilevel spinal segments with different numbers (0, 1, 2, and 3) of vertebral levels augmented with polymethylmethacrylate. SUMMARY OF BACKGROUND DATA: The stiffness and strength of single-level polymethylmethacrylate augmentations in individual and multilevel vertebrae treated by kyphoplasty and vertebroplasty have been studied, but the biomechanics of multilevel segments with more than 1 vertebral level augmented with polymethylmethacrylate are lacking, yet this is clinically relevant in multilevel compression fracture treatment. MATERIALS AND METHODS: A total of 48 multilevel segments (T3-T5, T6-T8, T9-T11, T12-L2, and L3-L5) from 12 spines with known bone mineral density (BMD) were allocated into 6 groups based on the number of vertebral levels augmented: 0 levels (n = 13), control group; 1 level (n = 7), group 2; 2 levels, groups 3, 4, and 5 (n = 7 in each); and 3 levels (n = 7), group 6. They were compressed to failure, disarticulated into individual vertebrae, and retested. Stiffness and strength were statistically analyzed using a univariate analysis of variance comparing the main effects, using least significant difference comparisons with 0.05 probability level. RESULTS: Strength was dependent on BMD (P < 0.001 multilevel segments, P < 0.001 individual vertebrae), with no differences among the 6 different augmentation groups, and no significant differences between augmented and nonaugmented individual vertebrae. Stiffness was dependent on BMD (P = 0.009 multilevel segments, P < 0.004 individual vertebrae), with no significant differences among the 6 different augmentation groups, and no significant differences between augmented and nonaugmented individual vertebrae. CONCLUSIONS: Multilevel segment biomechanics are dependent on BMD and not the pattern of augmentation, so the augmentation of fractured vertebrae can be extended to adjacent levels at risk for fracture to maintain stiffness and strength, thus preventing further fractures.

Aged↗
PubMed2006

Symptomatic cervical disc herniation in a pediatric Klippel-Feil patient: the risk of neural injury associated with extensive congenitally fused vertebrae and a hypermobile segment.

STUDY DESIGN: A case report. OBJECTIVE: To report the occurrence of a herniated cervical disc following a high-impact activity in a pediatric Klippel-Feil patient who presented with spinal cord compression, myelopathy, and myelomalacia requiring posterior instrumented fusion followed by an anterior discectomy and fusion at the hypermobile nonfused segment. SUMMARY OF BACKGROUND DATA: The primary hallmark of Klippel-Feil syndrome (KFS) is the presence of at least one congenitally fused cervical segment. Studies have reported the potential risk of cervical injury from hypermobility associated with the nonfused cervical segment in KFS. The manifestation of a cervical disc herniation in the pediatric KFS patient is rare. To the authors' knowledge, the development of a symptomatic cervical herniated disc attributed to mechanical fatigue following a high-impact activity has not been addressed in the literature with respect to the pediatric KFS patient having extensive cervical fusion and a hypermobile segment. METHODS: A 16.8-year-old KFS boy with occipitalization of C1 and fusion of C2-C3 and C4-T1 presented with myelopathy, severe cord compression, and myelomalacia stemming from a left-sided herniated cervical disc at C3-C4 with onset following an 8-foot high rooftop jump. On radiographic evaluation, the patient's C3-C4 segment was hypermobile. RESULTS: The patient was operatively managed via a same-day combined posterior-anterior procedure. The posterior aspect of the procedure entailed a posterior lateral mass plate-screw fixation at C3-C4 with autologous iliac crest bone fusion. Anteriorly, a discectomy was performed at C3-C4 with application of an interbody tricortical autograft. After surgery, the patient wore a halo vest for 3 months, followed by a soft collar for an additional 3 months. On final follow-up at 39 months, the patient was asymptomatic with no instrumentation-related complications, fusion of the posterior graft-bed and anterior interbody graft was noted, and cervical alignment was maintained. CONCLUSIONS: A hypermobile segment in the pediatric KFS patient is a risk factor that may lead to cord compression. A symptomatic herniated cervical disc may develop from an excessive mechanical load stress in a pediatric KFS patient with multiple fused segments. In such a patient, a same-day combined posterior-anterior procedure provides cord decompression and stabilizes the spine with a favorable outcome.

Adolescent↗
PubMed2003

Aerosol transfer of bladder urothelial and smooth muscle cells onto demucosalized colonic segments: a pilot study.

PURPOSE: We developed a cell transfer technology for covering demucosalized colonic segments with bladder urothelium. This covering would be achieved through aerosol spraying of single cell suspension of bladder urothelial and smooth muscle cells with fibrin glue onto the demucosalized colonic segments. MATERIALS AND METHODS: In 6 piglets (20 kg.) a 4 cm.2 area of bladder was excised. Single cell suspension of bladder urothelial and smooth muscle cells was prepared. A segment of detubularized sigmoid colon was isolated on its vascular pedicle and demucosalized. The single cell suspensions were combined with an equal volume of fibrin glue and sprayed over the raw submucosal surface of the sigmoid segment. The sigmoid segment was retubularized and sutured to the posterior peritoneum. Animals were sacrificed 4 weeks later, and the segment was submitted to histological and immunohistochemical analysis. RESULTS: Sigmoid segments appeared grossly intact with no reduction in surface area. Hematoxylin and eosin architecture revealed an intact urothelial layer. Deep to this layer was a randomly aligned but distinctly segregated layer of smooth muscle cells. The urological new smooth muscle layer stained positive for calponin and the urothelial layer was cytokeratin-7 and uroplakin III positive. CONCLUSIONS: Separation, cell suspension and aerosol delivery of bladder urothelial and smooth muscle cells in fibrin glue can successfully transfer these urological cell populations to a new host tissue commonly used in urological reconstruction. In vivo co-culture of bladder smooth muscle and urothelial cells results in coverage of a large area of demucosalized gut providing new potential for transfer and reconstitution of urologically functionally appropriate tissue to the bladder itself.

Aerosols↗
PubMed2004

Use of single layer small intestinal submucosa for long segment ureteral replacement: a pilot study.

PURPOSE: Previous studies have demonstrated successful use of small intestinal submucosa (SIS) as a tube for replacing short segment (11 mm) proximal ureteral defects. However, such small segment ureteral defects could be managed by resection re-anastomosis. We evaluated the use of 1-layer SIS as a tube for the replacement of long segment ureteral defects. MATERIALS AND METHODS: The ureters of 5 female mongrel dogs were accessed through a median laparotomy incision. A 4 cm segment of mid ureter was resected on the right side. The right ureteral segments were replaced by tubularized SIS segments using 6-zero polydioxanone interrupted sutures. Internal pigtail stents were left for 6 weeks. All animals were sacrificed at 12 weeks. Ureteral patency was assessed by excretory urography and magnetic resonance urography 7 and 12 weeks after the initial procedures. Inflammation and regeneration were assessed histologically. RESULTS: At 12 weeks all ureters on the experimental side were completely occluded with significant hydroureteronephrosis and the subsequent deterioration of kidney function. At autopsy there was failure to calibrate any of the experimental ureters with a 3Fr catheter. Although histologically urothelium and muscular cells had proliferated over the graft, they were embedded in an intense fibrotic and inflammatory process. CONCLUSIONS: Technically 1-layer SIS was easily modeled, providing the conditions for watertight anastomosis. The regeneration of urothelium and muscle was induced and supported by the graft. However, functional replacement was not successful. One-layer SIS is not a suitable material for replacing long segment (4 cm) ureteral defects.

Animals↗
PubMed1985

Cell lineage and segmentation in the leech.

Segments in the leech arise by the proliferation of longitudinally arrayed bandlets of blast cells derived from ten identifiable embryonic stem cells, two M, two N, four O/P and two Q teloblasts. In each bandlet, older blast cells lie ahead of those born later. By using microinjected cell lineage tracers it was shown previously that the teloblasts give rise to characteristic cell patterns made up of segmentally iterated complements of progeny designated as M, N, O, P and Q kinship groups. When a teloblast is injected after it has begun generating blast cells, a boundary is observed later in development between anterior, unlabelled progeny of blast cells produced before injection and posterior, labelled progeny of blast cells produced after injection. We have examined such boundaries in detail to establish the precise relationship between blast cell clones and segments, with the following conclusions: (i) in the M, O and P cell lines, one blast cell generates one segmental complement of progeny, but serially homologous blast clones intermix so that no segment boundaries can be defined based on primary blast cell clones; (ii) in the N and Q cell lines, two blast cells are required to generate a complete segmental complement of progeny; (iii) in the process of forming the germinal plate, cells derived from the N and Q teloblasts move past those derived from the M and O/P teloblasts, so that consegmental blast cell clones do not come into register until well after the establishment of segmentally iterated units within each bandlet.

Animals↗
PubMed1997

Complete nucleotide sequence of Colorado tick fever virus segments M6, S1 and S2.

The nucleotide sequences of the tenth (M6), eleventh (S1) and twelfth (S2) dsRNA genomic segments of the Florio strain (N-7180) of Colorado tick fever virus were determined and found to be 675, 998 and 1884 bp, respectively, in length. A nonanucleotide motif and a hexanucleotide motif were found to be highly conserved in the 5' and 3' non-coding regions (NCRs), respectively, of the three segments. The first and last three nucleotides of each segment were of inverted complementarity, and segment-specific inverted terminal repeats were detected in the NCRs of the three segments. These findings suggest the occurrence of intracellular panhandle structures for the RNA transcripts. A readthrough phenomenon is suspected in segment M6. The environment surrounding the opal codon (position 1052-1054) of segment M6 conforms to that of leaky opal codons described in the literature.

Animals↗
PubMed2002

Characterization of Campoletis sonorensis ichnovirus segment I genes as members of the repeat element gene family.

Campoletis sonorensis ichnovirus (CsIV) is a symbiotic virus associated with the endoparasitic wasp C. sonorensis. The virus is injected into the wasp's host, Heliothis virescens, during oviposition. One CsIV gene has been identified as a repeat element (rep) gene and encodes a ubiquitous imperfectly conserved 540 bp sequence. We report the sequencing and mapping of a rep-containing segment, segment I, that hybridizes to a known rep sequence from segment O(1). Analysis of this 8.6 kbp segment identified three ORFs having high similarity to the 540 bp rep sequence. All three rep sequence ORFs were expressed in parasitized H. virescens as well as in C. sonorensis tissues. Two of these rep genes, I 0.9 and I 1.1, have single copies of the 540 bp repeat sequence, while the third rep gene, I 1.2, has two imperfect copies, which are more similar to each other than to sequences on the segment I single-motif genes. Like the CsIV BHv 0.9 rep gene, the segment I rep genes lack introns and a signal peptide, suggesting that they are not secreted. Based on their similarity in nucleotide sequence, predicted amino acid sequence and gene structure, the three segment I repeat-containing genes, I 0.9, I 1.1 and I 1.2, are new members of the rep gene family.

Amino Acid Sequence↗
PubMed2004

Analysis of the medium (M) segment sequence of Guaroa virus and its comparison to other orthobunyaviruses.

Guaroa virus (GROV), a segmented virus in the genus Orthobunyavirus, has been linked to the Bunyamwera serogroup (BUN) through cross-reactivity in complement fixation assays of S segment-encoded nucleocapsid protein determinants, and also to the California serogroup (CAL) through cross-reactivity in neutralization assays of M segment-encoded glycoprotein determinants. Phylogenetic analysis of the S-segment sequence supported a closer relationship to the BUN serogroup for this segment and it was hypothesized that the serological reaction may indicate genome-segment reassortment. Here, cloning and sequencing of the GROV M segment are reported. Sequence analysis indicates an organization similar to that of other orthobunyaviruses, with genes in the order GN-NSm-Gc, and mature proteins generated by protease cleavage at one, and by signalase at possibly three, sites. A potential role of motifs that are more similar to CAL than to BUN virus sequences with respect to the serological reaction is discussed. No discernable evidence for reassortment was identified.

Amino Acid Sequence↗
PubMed2007

In vitro integration of an ichnovirus genome segment into the genomic DNA of lepidopteran cells.

Polydnaviruses (PDVs) are dsDNA viruses transmitted by ichneumonid and braconid endoparasitoids to their lepidopteran hosts during oviposition. Wasp carriers are asymptomatic and transmit the virus to their progeny through the germ line; replication is confined to the calyx region of the wasp ovary, where the virus accumulates in the fluid bathing the eggs. In the lepidopteran host, however, no virus replication takes place, but PDV gene expression is essential for successful parasitism. Sustained gene expression in the absence of virus replication thus requires that the circular PDV genome segments persist for days within host cells. Available evidence suggests that most genome segments persist as episomes, but recent studies have indicated that some genome segments may undergo integration within lepidopteran genomic DNA, at least in vitro. In the present study, an integrated form of a Tranosema rostrale ichnovirus (TrIV) genome segment was cloned from genomic DNA extracted from infected Choristoneura fumiferana CF-124T cells and junction regions on either side of the viral DNA sequence were sequenced. This is the first proven example of integration of an ichnovirus genome segment in infected lepidopteran cells. Interestingly, circular forms of this genome segment do not appear to persist in these cells; none the less, a gene (TrFrep1) carried by this genome segment displays long-term transcription in infected cultured cells.

Animals↗
PubMed2006

Structural investigation of disordered stress proteins. Comparison of full-length dehydrins with isolated peptides of their conserved segments.

Dehydrins constitute a class of intrinsically disordered proteins that are expressed under conditions of water-related stress. Characteristic of the dehydrins are some highly conserved stretches of seven to 17 residues that are repetitively scattered in their sequences, the K-, S-, Y-, and Lys-rich segments. In this study, we investigate the putative role of these segments in promoting structure. The analysis is based on comparative analysis of four full-length dehydrins from Arabidopsis (Arabidopsis thaliana; Cor47, Lti29, Lti30, and Rab18) and isolated peptide mimics of the K-, Y-, and Lys-rich segments. In physiological buffer, the circular dichroism spectra of the full-length dehydrins reveal overall disordered structures with a variable content of poly-Pro helices, a type of elongated secondary structure relying on bridging water molecules. Similar disordered structures are observed for the isolated peptides of the conserved segments. Interestingly, neither the full-length dehydrins nor their conserved segments are able to adopt specific structure in response to altered temperature, one of the factors that regulate their expression in vivo. There is also no structural response to the addition of metal ions, increased protein concentration, or the protein-stabilizing salt Na(2)SO(4). Taken together, these observations indicate that the dehydrins are not in equilibrium with high-energy folded structures. The result suggests that the dehydrins are highly evolved proteins, selected to maintain high configurational flexibility and to resist unspecific collapse and aggregation. The role of the conserved segments is thus not to promote tertiary structure, but to exert their biological function more locally upon interaction with specific biological targets, for example, by acting as beads on a string for specific recognition, interaction with membranes, or intermolecular scaffolding. In this perspective, it is notable that the Lys-rich segment in Cor47 and Lti29 shows sequence similarity with the animal chaperone HSP90.

Amino Acid Sequence↗
PubMed2004

Combining strings and necklaces for interactive three-dimensional segmentation of spinal images using an integral deformable spine model.

Segmentation of the spine directly from three-dimensional (3-D) image data is desirable to accurately capture its morphological properties. We describe a method that allows true 3-D spinal image segmentation using a deformable integral spine model. The method learns the appearance of vertebrae from multiple continuous features recorded along vertebra boundaries in a given training set of images. Important summarizing statistics are encoded into a necklace model on which landmarks are differentiated on their free dimensions. The landmarks are used within a priority segmentation scheme to reduce the complexity of the segmentation problem. Necklace models are coupled by string models. The string models describe in detail the biological variability in the appearance of spinal curvatures from multiple continuous features recorded in the training set. In the segmentation phase, the necklace and string models are used to interactively detect vertebral structures in new image data via elastic deformation reminiscent of a marionette with strings allowing for movement between interrelated structures. Strings constrain the deformation of the spine model within feasible solutions. The driving application in this work is analysis of computed tomography scans of the human lumbar spine. An illustration of the segmentation process shows that the method is promising for segmentation of the spine and for assessment of its morphological properties.

Aged↗
PubMed2005

An image model and segmentation algorithm for reflectance confocal images of in vivo cervical tissue.

The automatic segmentation of nuclei in confocal reflectance images of cervical tissue is an important goal toward developing less expensive cervical precancer detection methods. Since in vivo confocal reflectance microscopy is an emerging technology for cancer detection, no prior work has been reported on the automatic segmentation of in vivo confocal reflectance images. However, prior work has shown that nuclear size and nuclear-to-cytoplasmic ratio can determine the presence or extent of cervical precancer. Thus, segmenting nuclei in confocal images will aid in cervical precancer detection. Successful segmentation of images of any type can be significantly enhanced by the introduction of accurate image models. To enable a deeper understanding of confocal reflectance microscopy images of cervical tissue, and to supply a basis for parameter selection in a classification algorithm, we have developed a model that accounts for the properties of the imaging system and of the tissues. Using our model in conjunction with a powerful image enhancement tool (anisotropic median-diffusion), appropriate statistical image modeling of spatial interactions (Gaussian Markov random fields), and a Bayesian framework for classification-segmentation, we have developed an effective algorithm for automatically segmenting nuclei in confocal images of cervical tissue. We have applied our algorithm to an extensive set of cervical images and have found that it detects 90% of hand-segmented nuclei with an average of 6 false positives per frame.

Algorithms↗
PubMed2006

Range image segmentation using surface selection criterion.

In this paper, we address the problem of recovering the true underlying model of a surface while performing the segmentation. First, and in order to solve the model selection problem, we introduce a novel criterion, which is based on minimising strain energy of fitted surfaces. We then evaluate its performance and compare it with many other existing model selection techniques. Using this criterion, we then present a robust range data segmentation algorithm capable of segmenting complex objects with planar and curved surfaces. The presented algorithm simultaneously identifies the type (order and geometric shape) of each surface and separates all the points that are part of that surface. This paper includes the segmentation results of a large collection of range images obtained from objects with planar and curved surfaces. The resulting segmentation algorithm successfully segments various possible types of curved objects. More importantly, the new technique is capable of detecting the association between separated parts of a surface, which has the same Cartesian equation while segmenting a scene. This aspect is very useful in some industrial applications of range data analysis.

Algorithms↗
PubMed2006

Image and texture segmentation using local spectral histograms.

We present a method for segmenting images consisting of texture and nontexture regions based on local spectral histograms. Defined as a vector consisting of marginal distributions of chosen filter responses, local spectral histograms provide a feature statistic for both types of regions. Using local spectral histograms of homogeneous regions, we decompose the segmentation process into three stages. The first is the initial classification stage, where probability models for homogeneous texture and nontexture regions are derived and an initial segmentation result is obtained by classifying local windows. In the second stage, we give an algorithm that iteratively updates the segmentation using the derived probability models. The third is the boundary localization stage, where region boundaries are localized by building refined probability models that are sensitive to spatial patterns in segmented regions. We present segmentation results on texture as well as nontexture images. Our comparison with other methods shows that the proposed method produces more accurate segmentation results.

Algorithms↗
PubMed2005

MRI fuzzy segmentation of brain tissue using neighborhood attraction with neural-network optimization.

Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of magnetic resonance (MR) images. Unfortunately, MR images always contain a significant amount of noise caused by operator performance, equipment, and the environment, which can lead to serious inaccuracies with segmentation. A robust segmentation technique based on an extension to the traditional fuzzy c-means (FCM) clustering algorithm is proposed in this paper. A neighborhood attraction, which is dependent on the relative location and features of neighboring pixels, is shown to improve the segmentation performance dramatically. The degree of attraction is optimized by a neural-network model. Simulated and real brain MR images with different noise levels are segmented to demonstrate the superiority of the proposed technique compared to other FCM-based methods. This segmentation method is a key component of an MR image-based classification system for brain tumors, currently being developed. Index Terms-Improved fuzzy c-means clustering (IFCM), magnetic resonance imaging (MRI), neighborhood attraction, segmentation.

Algorithms↗