Estimation of Depth Map Using Image Focus: A Scale-Space Approach for Shape Recovery

4-D Hippocampus Measurements of Neurite Growth in vitro

“…An important requirement of the algorithms developed for this work is to accurately determine within the 2-D field of view (FOV) the dynamic depth positions of the individual neurites in the z-plane over the course of each experiment [4]. Here, images of the cells were captured using automated microscopy with a high precision Z-stage (0.4 μm), from which a series of 89 images were focused at different heights, collectively forming a so-called z-stack of images.…”

“…Here, images of the cells were captured using automated microscopy with a high precision Z-stage (0.4 μm), from which a series of 89 images were focused at different heights, collectively forming a so-called z-stack of images. The zposition of the individual cell/feature can then be estimated using the multi-scale focus estimation procedure described in [4]. The cell organelle are analysed by means of the most in focus pixels whereby their dynamic locations (x,y,z,t) can be determined.…”

“…This simple 2-D, laminar circuitry provides an optimum target for in vitro modelling of the 3-D structure of functional neurones and extensive synaptic networks. However, previous work on in vitro hippocampal networks have largely focused on 2-D layers, with little evidence to support extrapolation of findings to 3-D constructs and, more significantly, no evidence of 3-D constructs formed of laminar 2-D hippocampal networks [4]. Using a laminar 2-D network of orthogonally orientated nanofibres, where complex cellular studies were carried out in vitro within unlabelled continuous live cultures, this paper presents our current work in progress on analysing such a 3-D model of the hippocampal circuitry.…”

“…Specifically, to help investigate the properties of primary embryonic hippocampal neurons in the presence of various cues within the 3-D scaffold, this study introduced 3-D analysis of in vitro neurological structures which have only been examined in 2-D to date. The work built on our earlier research on 3-D shape/depth estimation [4] that utilised time lapse videos to explore in detail changes in neurite growth. Here, the shape established by a focus/defocus algorithm was developed to provide an accurate representation of the biological growth within the 3-D structure [5].…”

Label free cell tracking in 3D tissue engineering constructs with high resolution imaging

“…In this study, the object is projected into defocused images depending on the distance from the lens objective. These distances influence the observation of details on different scales; the Gaussian blurring causes a greater loss of information on local features such as the cell membrane, than on global features such as the cultures general topography [10,11]. This blurring increases proportionately with the objects distance from the depth-of-view of the lens.…”