Causes of the apparent scale independence of fractal indices associated with forest fragmentation in Bolivia

In a recent investigation, Landsat TM and ETM+ data were used to simulate different resolutions of remotely-sensed images (from 30 to 1100 m) and to analyze the effect of resolution on a range of landscape metrics associated with spatial patterns of forest fragmentation in Chapare, Bolivia since the mid-1980s. Whereas most metrics were found to be highly dependent on pixel size, several fractal metrics (DLFD, MPFD, and AWMPFD) were apparently independent of image resolution, in contradiction with a sizeable body of literature indicating that fractal dimensions of natural objects depend strongly on image characteristics. The present re-analysis of the Chapare images, using two alternative algorithms routinely used for the evaluation of fractal dimensions, shows that the values of the box-counting and information fractal dimensions are systematically larger, sometimes by as much as 85%, than the "fractal" indices DLFD, MPFD, and AWMFD for the same images. In addition, the geometrical fractal features of the forest and non-forest patches in the Chapare region strongly depend on the resolution of images used in the analysis. The largest dependency on resolution occurs for the box-counting fractal dimension in the case of the non-forest patches in 1993, where the difference between the 30 and I 100 m-resolution images corresponds to 24% of the full theoretical range (1.0 to 2.0) of the mass fractal dimension. The observation that the indices DLFD, MPFD, and AWMPFD, unlike the classical fractal dimensions, appear relatively unaffected by resolution in the case of the Chapare images seems due essentially to the fact that these indices are based on a heuristic, "non-geometric" approach to fractals. Because of their lack of a foundation in fractal geometry, nothing guarantees that these indices will be resolution-independent in general. (C) 2006 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved.

Activated mast cells in proximity to colonic nerves correlate with abdominal pain in irritable bowel syndrome

BACKGROUND & AIMS: The mechanisms underlying abdominal pain perception in irritable bowel syndrome (IBS) are poorly understood. Intestinal mast cell infiltration may perturb nerve function leading to symptom perception. We assessed colonic mast cell infiltration, mediator release, and spatial interactions with mucosal innervation and their correlation with abdominal pain in IBS patients. METHODS: IBS patients were diagnosed according to Rome II criteria and abdominal pain quantified according to a validated questionnaire. Colonic mucosal mast cells were identified immunohistochemically and quantified with a computer-assisted counting method. Mast cell tryptase and histamine release were analyzed immunoenzymatically. Intestinal nerve to mast cell distance was assessed with electron microscopy. RESULTS: Thirty-four out of 44 IBS patients (77%) showed an increased area of mucosa occupied by mast cells as compared with controls (9.2% +/- 2.5% vs. 3.3 +/- 0.8%, respectively; P < 0.001). There was a 150% increase in the number of degranulating mast cells (4.76 +/- 3.18/field vs. 2.42 +/- 2.26/field, respectively; P = 0.026). Mucosal content of tryptase was increased in IBS and mast cells spontaneously released more tryptase (3.22 +/- 3.48 pmol/min/mg vs. 0.87 +/- 0.65 pmol/min/mg, respectively; P = 0.015) and histamine (339.7 +/- 59.0 ng/g vs. 169.3 +/- 130.6 ng/g, respectively; P = 0.015). Mast cells located within 5 microm of nerve fibers were 7.14 +/- 3.87/field vs. 2.27 +/- 1.63/field in IBS vs. controls (P < 0.001). Only mast cells in close proximity to nerves were significantly correlated with severity and frequency of abdominal pain/discomfort (P < 0.001 and P = 0.003, respectively). CONCLUSIONS: Colonic mast cell infiltration and mediator release in proximity to mucosal innervation may contribute to abdominal pain perception in IBS patients.

A novel method for measuring lag times in division of individual bacterial cells using image analysis

A method is presented for determining the time to first division of individual bacterial cells growing on agar media. Bacteria were inoculated onto agar-coated slides and viewed by phase-contrast microscopy. Digital images of the growing bacteria were captured at intervals and the time to first division estimated by calculating the "box area ratio". This is the area of the smallest rectangle that can be drawn around an object, divided by the area of the object itself. The box area ratios of cells were found to increase suddenly during growth at a time that correlated with cell division as estimated by visual inspection of the digital images. This was caused by a change in the orientation of the two daughter cells that occurred when sufficient flexibility arose at their point of attachment. This method was used successfully to generate lag time distributions for populations of Escherichia coli, Listeria monocytogenes and Pseudomonas aeruginosa, but did not work with the coccoid organism Staphylococcus aureus. This method provides an objective measure of the time to first cell division, whilst automation of the data processing allows a large number of cells to be examined per experiment. (c) 2005 Elsevier B.V. All rights reserved.

Robust background model for pixel based people counting using a single uncalibrated camera

Several pixel-based people counting methods have been developed over the years. Among these the product of scale-weighted pixel sums and a linear correlation coefficient is a popular people counting approach. However most approaches have paid little attention to resolving the true background and instead take all foreground pixels into account. With large crowds moving at varying speeds and with the presence of other moving objects such as vehicles this approach is prone to problems. In this paper we present a method which concentrates on determining the true-foreground, i.e. human-image pixels only. To do this we have proposed, implemented and comparatively evaluated a human detection layer to make people counting more robust in the presence of noise and lack of empty background sequences. We show the effect of combining human detection with a pixel-map based algorithm to i) count only human-classified pixels and ii) prevent foreground pixels belonging to humans from being absorbed into the background model. We evaluate the performance of this approach on the PETS 2009 dataset using various configurations of the proposed methods. Our evaluation demonstrates that the basic benchmark method we implemented can achieve an accuracy of up to 87% on sequence ¿S1.L1 13-57 View 001¿ and our proposed approach can achieve up to 82% on sequence ¿S1.L3 14-33 View 001¿ where the crowd stops and the benchmark accuracy falls to 64%.