Clinical Center

Many types of glial-neuronal interactions occur during the development of the nervous system. To determine how such interactions might affect the development of autonomic ganglia, we compared the morphology of embryonic rat sympathetic neurons grown in the absence and in the presence of ganglionic nonneuronal cells in serum-free medium. Dye injections, electron microscopy, and immunocytochemistry were used to distinguish axons from dendrites. In cultures without nonneuronal cells, most (<80%) sympathetic neurons extended only a single axonal process, and this unipolar state persisted for at least 8 weeks. Coculture with ganglionic nonneuronal cells caused sympathetic neurons to become multipolar and to extend multiple (range 1–17) dendrites. Morphometric measurements made after 1 month of coculture indicated that the amount of dendritic growth that occurred in vitro (mean number of dendrites/cell = 7.5; total dendritic length = 1,050 μm) was similar to that normally occurring during a comparable period in situ. In contrast to its prominent effects on dendritic growth, coculture did not cause chages in the number of axons/neuron or in the uptake of neurotransmitter. Cultures with ganglionic nonneuronal cells were immunostained for antigens present on the surfaces of fibroblasts (Thy-1.1, fibronectin) and of glia of the peripheral nervous system (laminin). Fewer than 1% of the nonneuronal cells displayed immunoreactivity for fibroblastic antigens; in contrast, ≥99% reacted with antibody to laminin. Moreover, reconstitution experiments revealed that purified populations of laminin-positive Schwann cells promoted dendritic growth. Fibroblasts and heart cells lacked this activity. These data indicate that glia selectively promote dendritic development in sympathetic neurons maintained in serum-free medium.

Inflammation-promoting activity of HMGB1 on human microvascular http://bloodjournal.hematologylibrary.org/content/101/7/2652.full.html Updated information and services can be found at: (2497 articles) Hemostasis, Thrombosis, and Vascular Biology Articles on similar topics can be found in the following Blood collections http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests

Macrophage inflammatory protein (MIP)-1a and MIP-1b regulate leukocyte activation and trafficking. To assess the role of MIP-1a and MIP-1b in human inflammation, healthy subjects were studied during experimental endotoxemia with prior administration of ibuprofen, a cyclooxygenase inhibitor, or dimeric p75 tumor necrosis factor (TNF)-a receptor, a TNF antagonist; septic patients were also studied. Following endotoxin, blood levels of both MIP-1 molecules rose acutely and fell to baseline by 6 h ( ). While MIP-1 mediates fever P ϭ .001 in animals independent of cyclooxygenase blockade, in subjects given endotoxin and ibuprofen, MIP-1 levels increased and fever was suppressed. MIP-1 levels were not diminished by inhibiting circulating TNF-a in humans. In septic patients, elevated levels of MIP-1a and MIP-1b were detected within 24 h of sepsis and fell in parallel with TNF-a and interleukin-6 ( ). MIP-1a and MIP-1b increase during acute inflammation but are not associated P ! .01 with fever in endotoxemic humans during cyclooxygenase blockade.

Colonography is an important screening tool for colorectal lesions. This paper presents a method for establishing spatial correspondence between prone and supine inner colon surfaces reconstructed from CT colonography. The method is able to account for the large deformations and torsions of the colon occurring through movement between the two positions. Therefore, we parameterised the two surfaces in order to provide a 2D indexing system over the full length of the colon using the Ricci flow method. This provides the input to a non-rigid B-spline registration in 2D space which establishes a correspondence for each surface point of the colon in prone and supine position. The method was validated on twelve clinical cases and demonstrated promising registration results over the majority of the colon surface. Additional material can be found at http://cmic.cs.ucl.ac.uk/staff/mingxing_hu/endoscopic_images/ .

Purpose: Computed tomographic (CT) colonography is a relatively new technique for detecting bowel cancer or potentially precancerous polyps. CT scanning is combined with 3-dimensional image reconstruction to produce a virtual endoluminal representation similar to optical colonoscopy. Because retained fluid and stool can mimic pathology, CT data is acquired with the bowel cleansed and insufflated with gas and patient in both prone and supine positions. Radiologists then match visually endoluminal locations between the two acquisitions in order to determine whether apparent pathology is real or not. This process is hindered by the fact that the colon, essentially a long tube, can undergo considerable deformation between acquisitions. We present a novel approach to automatically establish spatial correspondence between prone and supine endoluminal colonic surfaces after surface parameterization, even in the case of local colon collapse.

CT colonography is routinely performed with the patient prone and supine to differentiate fixed colonic pathology from mobile faecal residue. We propose a novel method to automatically establish correspondence. Haustral folds are detected using a graph cut method applied to a surface curvature-based metric, where image patches are generated using endoluminal CT colonography surface rendering. The intensity difference between image pairs, along with additional neighbourhood information to enforce geometric constraints, are used with a Markov Random Field (MRF) model to estimate the fold labelling assignment. The method achieved fold matching accuracy of 83.1% and 88.5% with and without local colonic collapse. Moreover, it improves an existing surface-based registration algorithm, decreasing mean registration error from 9.7mm to 7.7mm in cases exhibiting collapse.

This paper provides an external validation of a prone-supine registration algorithm for CT colonography (CTC). A validation sample of 49 patient cases with 66 polyps (6 to 30 mm) was selected from a publicly available, anonymized CTC archive. To enhance generalizability, no case was excluded due to poor preparation or inadequate distension. Corresponding prone and supine polyp coordinates were recorded and the endoluminal surfaces registered: a Markov Random Field technique was used to find feature matches between prone/supine acquisitions and following mapping of the endoluminal surface to a cylinder, dense surface correspondence was achieved via cylindrical non-rigid registration. The polyp registration error was determined and a subjective assessment of registration made for 2D slice-based and 3D endoluminal data display using a pre-specified scoring system. Results were compared to using “normalized distance along the colon centerline” (NDACC) which approximates to the method currently employed to match colonic positions using proprietary CT colonography interpretation software. Registration was possible in all 49 cases. Overall mean 3D polyp registration error was significantly smaller with 19.9 mm in comparison to 27.7 mm using NDACC (p=0.001). 82.7% of polyp matches were defined as “successful” in comparison to 37.1% using NDACC according to the pre-specified criteria. Similarly, using 2D visualization, 62.1% registrations were “successful” and only 22.7% using NDACC. Full surface-based prone-to-supine registration can successfully map the location of a polyp identified on one acquisition to the corresponding endoluminal surface in the opposing acquisition, greatly facilitating polyp matching and aiding interpretation. Our method compares favorably to using NDACC.

CT colonography interpretation is difficult and time-consuming because fecal residue or fluid can mimic or obscure polyps, leading to diagnostic errors. To compensate for this, it is normal practice to obtain CT data with the patient in prone and supine positions. Repositioning redistributes fecal residue and colonic gas; fecal residue tends to move, while fixed mural pathology does not.

CT-Colonography is widely utilised in the diagnosis of colorectal cancer. However, the technique is difficult and time-consuming because residual stool may simulate colorectal cancer or potentially precancerous polyps. Consequently, scans are routinely performed with the patient in both prone and supine positions to help differentiate mobile faecal matter from fixed abnormalities of the colonic wall. Currently, the prone and supine datasets have to be manually aligned which can be very difficult due to the significant colonic deformation which takes place during patient movement. Establishing accurate spatial correspondence between prone and supine inner colon surfaces reconstructed from CT images can therefore potentially improve the accuracy and reduce the time taken for interpretation. This paper presents a method for mapping the inner colonic surface to a cylinder using the Ricci flow which provides a 2D indexing system over the full length of the colon. This procedure converts the registration task from a 3D to a 2D problem and could help cope with the large deformations occurring between prone and supine positions.

Computed Tomographic (CT) colonography is a technique used for the detection of bowel cancer or potentially precancerous polyps. The procedure is performed routinely with the patient both prone and supine to differentiate fixed colonic pathology from mobile faecal residue. Matching corresponding locations is difficult and time consuming for radiologists due to colonic deformations that occur during patient repositioning.

This invited presentation summarizes recent advances in the incorporation of knowledge of the geometry, tissue mechanical properties and imaging characteristics in establishing spatial correspondence between multiple images of highly deforming, soft tissue structures. Spatial correspondence is used to aid diagnosis and in the extraction of quantitative parameters for disease detection, monitoring disease progression and assessing therapeutic response. The work is illustrated through clinical examples of multi-modal imaging of the breast, assessment of small bowel motility and polyp detection in the large bowel.

Enlarged lymph nodes (LNs) can provide important information for cancer diagnosis, staging, and measuring treatment reactions, making automated detection a highly sought goal. In this paper, we propose a new algorithm representation of decomposing the LN detection problem into a set of 2D object detection subtasks on sampled CT slices, largely alleviating the curse of dimensionality issue. Our 2D detection can be effectively formulated as linear classification on a single image feature type of Histogram of Oriented Gradients (HOG), covering a moderate field-of-view of 45 by 45 voxels. We exploit both max-pooling and sparse linear fusion schemes to aggregate these 2D detection scores for the final 3D LN detection. In this manner, detection is more tractable and does not need to perform perfectly at instance level (as weak hypotheses) since our aggregation process will robustly harness collective information for LN detection. Two datasets (90 patients with 389 mediastinal LNs and 86 patients with 595 abdominal LNs) are used for validation. Cross-validation demonstrates 78.0% sensitivity at 6 false positives/volume (FP/vol.) (86.1% at 10 FP/vol.) and 73.1% sensitivity at 6 FP/vol. (87.2% at 10 FP/vol.), for the mediastinal and abdominal datasets respectively. Our results compare favorably to previous state-of-the-art methods.

Potentially precancerous polyps detected with CT colonography (CTC) need to be removed subsequently, using an optical colonoscope (OC). Due to large colonic deformations induced by the colonoscope, even very experienced colonoscopists find it difficult to pinpoint the exact location of the colonoscope tip in relation to polyps reported on CTC. This can cause unduly prolonged OC examinations that are stressful for the patient, colonoscopist and supporting staff.

Automated classification of human anatomy is an important prerequisite for many computer-aided diagnosis systems. The spatial complexity and variability of anatomy throughout the human body makes classification difficult. "Deep learning" methods such as convolutional networks (ConvNets) outperform other state-of-the-art methods in image classification tasks. In this work, we present a method for organ-or bodypart-specific anatomical classification of medical images acquired using computed tomography (CT) with ConvNets. We train a ConvNet, using 4,298 separate axial 2D key-images to learn 5 anatomical classes. Key-images were mined from a hospital PACS archive, using a set of 1,675 patients. We show that a data augmentation approach can help to enrich the data set and improve classification performance. Using Conv-Nets and data augmentation, we achieve anatomy-specific classification error of 5.9 % and area-under-the-curve (AUC) values of an average of 0.998 in testing. We demonstrate that deep learning can be used to train very reliable and accurate classifiers that could initialize further computer-aided diagnosis.

Automated computer-aided detection (CADe) in medical imaging has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities but typically with the cost of high false-positives (FP) rates per patient. We design a two-tiered coarse-to-fine cascade framework to first operate a highly sensitive candidate generation system at a maximum sensitivity of ∼100% but with high FP level (∼50 per patient), leveraging the existing CAD systems. Regions or volumes of interest (ROI or VOI) for lesion candidates are generated in this step and function as input for a second tier which is our focus. In this stage, we generate N 2D (dimensional) or 2.5D views, via sampling through scale transformations, random translations and rotations with respect to each ROI centroid coordinates. These random views are used to train deep Convolutional Neural Network (ConvNet) classifiers. In testing, the trained ConvNets are employed to assign class (e.g., lesion, pathology) probabilities for a new set of N random views that are averaged then at each ROI to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities in three different data sets: 59 patients for sclerotic metastases detection, 176 for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to varying sizes of data sets. Our proposed method improves the CADe performance markedly in all cases. CADe sensitivities improved from 57% to 70%, from 43% to 77% and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively.

Enlarged lymph nodes (LNs) can provide important information for cancer diagnosis, staging, and measuring treatment reactions, making automated detection a highly sought goal. In this paper, we propose a new algorithm representation of decomposing the LN detection problem into a set of 2D object detection subtasks on sampled CT slices, largely alleviating the curse of dimensionality issue. Our 2D detection can be effectively formulated as linear classification on a single image feature type of Histogram of Oriented Gradients (HOG), covering a moderate field-of-view of 45 by 45 voxels. We exploit both simple pooling and sparse linear fusion schemes to aggregate these 2D detection scores for the final 3D LN detection. In this manner, detection is more tractable and does not need to perform perfectly at instance level (as weak hypotheses) since our aggregation process will robustly harness collective information for LN detection. Two datasets (90 patients with 389 mediastinal LNs and 86 patients with 595 abdominal LNs) are used for validation. Cross-validation demonstrates 78.0% sensitivity at 6 false positives/volume (FP/vol.) (86.1% at 10 FP/vol.) and 73.1% sensitivity at 6 FP/vol. (87.2% at 10 FP/vol.), for the mediastinal and abdominal datasets respectively. Our results compare favorably to previous state-of-the-art methods.

Robust automated organ segmentation is a prerequisite for computer-aided diagnosis (CAD), quantitative imaging analysis, detection of pathologies and surgical assistance. For anatomical high-variability organs such as the pancreas, previous segmentation approaches report low accuracies in comparison to well studied organs like the liver or heart. We present a fully-automated bottom-up approach for pancreas segmentation in abdominal computed tomography (CT) scans. The method is based on a hierarchical cascade of information propagation by classifying image patches at different resolutions and (segments) superpixels. There are four stages in the system: 1) decomposing CT slice images as a set of disjoint boundary-preserving superpixels; 2) computing pancreas class probability maps via dense patch labeling; 3) classifying superpixels by pooling both intensity and probability features to form empirical statistics in cascaded random forest frameworks; and 4) simple connectivity based post-processing. The dense image patch labeling is conducted by two schemes: efficient random forest classifier on image histogram, location and texture features; and more expensive (but with better specificity) deep convolutional neural network classification, on larger image windows (i.e., with more spatial contexts). Oversegmented 2D CT slices by the Simple Linear Iterative Clustering approach are adopted through model/parameter calibration and labeled at the superpxiel level for positive (pancreas) or negative (non-pancreas background) classes. Evaluation of the approach is done on a database of 80 manually segmented CT volumes in six-fold cross-validation. Our achieved results are comparable, or better than the state-of-the-art methods (evaluated by "leave-one-patient-out"), with a Dice coefficient of 70.7% and Jaccard Index of 57.9%. In addition, the computational efficiency has been drastically improved in the order of 6 ∼ 8 minutes, comparing with others of ≥ 10 hours per testing case. The segmentation framework using deep patch labeling confidences is also more numerically stable, reflected by the smaller performance metric standard deviations.