Anil A. Bharath

Retinal vascular tree morphology: a semi-automatic quantification

A semi-automatic method to measure and quantify geometrical and topological properties of continuous vascular trees in clinical fundus images is described. Measurements are made from binary images obtained with a previously described segmentation process. The skeletons of the segmented trees are produced by thinning, ff branch and crossing points are identified and segments of the trees are labeled and stored as a chain code. The operator selects a tree to be measured and decides if it is an arterial or venous tree. An automatic process then measures the lengths, areas and angles of the individual segments of the tree. Geometrical data and the connectivity information between branches from continuous retinal vessel trees are tabulated. A number of geometrical properties and topological indexes are derived. Vessel diameters and branching angles are validated against manual measurements and several derived geometrical and topological properties are extracted from red-free fundus images of ten normotensive and ten age- and sex-matched hypertensive subjects and compared with previously reported results.

Carotid artery wall motion estimated from b-mode ultrasound using region tracking and block matching

The motion of the carotid atheromatous plaque relative to the adjacent wall may be related to the risk of cerebral events. A quantitative method for motion estimation was applied to analyse arterial wall movement from sequences of 2-D B-mode ultrasound (US) images. Image speckle patterns were tracked between successive frames using the correlation coefficient as the matching criterion. The size of the selected region-of-interest (ROI) was shown to affect the motion analysis results; an optimal size of 3.2 x 2.5 mm(2) was suggested for tracking a region at the wall-lumen interface and of 6.3 x 2.5 mm(2) for one within the tissue. The results showed expected cyclical motion in the radial direction and some axial movement of the arterial wall. The method can be used to study further the axial motion of the carotid artery wall and plaque and, thus, provide useful insight into the mechanisms of atherosclerosis.

Retinal Blood Vessel Segmentation by Means of Scale-Space Analysis and Region Growing

We present a method for retinal blood vessel segmentation based upon the scale-space analysis of the first and second derivative of the intensity image which gives information about its topology and overcomes the problem of variations in contrast inherent in these images. We use the local maxima over scales of the magnitude of the gradient and the maximum principal curvature as the two features used in a region growing procedure. In the first stage, the growth is constrained to regions of low gradient magnitude. In the final stage this constraint is relaxed to allow borders between regions to be defined. The algorithm is tested in both red-free and fluorescein retinal images.

Computer algorithms for the automated measurement of retinal arteriolar diameters.

AIMS Quantification of retinal vascular change is difficult and manual measurements of vascular features are slow and subject to observer bias. These problems may be overcome using computer algorithms. Three automated methods and a manual method for measurement of arteriolar diameters from digitised red-free retinal photographs were compared. METHODS 60 diameters (in pixels) measured by manual identification of vessel edges in red-free retinal images were compared with diameters measured by (1) fitting vessel intensity profiles to a double Gaussian function by non-linear regression, (2) a standard edge detection algorithm (Sobel), and (3) determination of points of maximum intensity variation by a sliding linear regression filter (SLRF). Method agreement was analysed using Bland–Altman plots and the repeatability of each method was assessed. RESULTS Diameter estimations obtained using the SLRF method were the least scattered although diameters obtained were approximately 3 pixels greater than those measured manually. The SLRF method was the most repeatable and the Gaussian method less so. The Sobel method was the least consistent owing to frequent misinterpretation of the light reflex as the vessel edge. CONCLUSION Of the three automated methods compared, the SLRF method was the most consistent (defined as the method producing diameter estimations with the least scatter) and the most repeatable in measurements of retinal arteriolar diameter. Application of automated methods of retinal vascular analysis may be useful in the assessment of cardiovascular and other diseases.

A method of vessel tracking for vessel diameter measurement on retinal images

A method of vessel tracking has been developed for quantification of vessel diameters of retinal images. This method utilises twin Gaussian functions to model the distribution of grey level over a vessel cross section. The diameter of the vessel at the cross section can then be calculated using the functions. The variation of vessel diameter in the direction of vessel longitude axis has been described by a tracking technique based on parameters of modelled intensity distribution curves over every cross section. This enables us to obtain an average diameter over any length of a vessel and to develop more parameters for diagnosis and study of vascular diseases.

Quantification and characterisation of arteries in retinal images

A computerised system is presented for the automatic quantification of blood vessel topography in retinal images. This system utilises digital image processing techniques to provide more reliable and comprehensive information for the retinal vascular network. It applies strategies and algorithms for measuring vascular trees and includes methods for locating the centre of a bifurcation, detecting vessel branches, estimating vessel diameter, and calculating angular geometry at a bifurcation. The performance of the system is studied by comparison with manual measurements and by comparing measurements between red-free images and fluorescein images. In general an acceptable degree of accuracy and precision was seen for all measurements, although the system had difficulty dealing with very noisy images and small or especially tortuous blood vessels.

Segmentation of retinal blood vessels based on the second directional derivative and region growing

We present a method for the segmentation of blood vessels in retinal images based upon the second derivative of the intensity image which gives information about its topology and overcomes the problem of image intensity variations. The minimum eigenvalue and the magnitude of its gradient are used as features for a region growing procedure which is defined in two stages. For the first stage, growth is restricted to regions with low gradients, allowing vessels to grow where the values of the minimum eigenvalue lie within a wide interval and allowing rapid growth of background regions outside of the vessel boundaries. For the second stage, in which the borders between classes are defined, the algorithm grows vessel and background classes simultaneously without the gradient restriction.

Deep Reinforcement Learning: A Brief Survey

Deep reinforcement learning is poised to revolutionise the field of AI and represents a step towards building autonomous systems with a higher level understanding of the visual world. Currently, deep learning is enabling reinforcement learning to scale to problems that were previously intractable, such as learning to play video games directly from pixels. Deep reinforcement learning algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of reinforcement learning, then progress to the main streams of value-based and policybased methods. Our survey will cover central algorithms in deep reinforcement learning, including the deep Q-network, trust region policy optimisation, and asynchronous advantage actor-critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via reinforcement learning. To conclude, we describe several current areas of research within the field.Deep reinforcement learning (DRL) is poised to revolutionize the field of artificial intelligence (AI) and represents a step toward building autonomous systems with a higherlevel understanding of the visual world. Currently, deep learning is enabling reinforcement learning (RL) to scale to problems that were previously intractable, such as learning to play video games directly from pixels. DRL algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of RL, then progress to the main streams of value-based and policy-based methods. Our survey will cover central algorithms in deep RL, including the deep Q-network (DQN), trust region policy optimization (TRPO), and asynchronous advantage actor critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via RL. To conclude, we describe several current areas of research within the field.

A survey of architecture and function of the primary visual cortex (V1)

The largest visual area, known as the primary visual cortex or V1, has greatly contributed to the current understanding of mammalian and human visual pathways and their role in visual perception. The initial discovery of orientation-sensitive neurons in V1, arranged according to a retinotopic mapping, suggested an analogy to its function as a low-level feature analyzer. Subsequent discoveries of phase, spatial frequency, color, ocular origin, and direction-of-motion-sensitive neurons, arranged into overlapping maps, further lent support to the view that it performs a rich decomposition, similar to signal processing transforms, of the retinal output. Like the other cortical areas, V1 has a laminar organization with specialization for input from the relayed retinal afferents, output to the higher visual areas, and the segregation of the magno (motion) and parvo (form) pathways. Spatially lateral connections that exist between neurons of similar and varying properties have also been proposed to give rise to a computation of a bottom-up saliency map in V1. We provide a review of the selectivity of neurons in V1, laminar specialization and analogies to signal processing techniques, a model of V1 saliency computation, and higher-area feedback that may mediate perception.

Steerable Filters from Erlang Functions

Scale and orientation steerable 2D filters are constructed using a frame of Erlang functions in the Fourier domain. Erlang forms for the radial frequency characteristic are shown to provide complex quadrature filters which can be steered in a scale parameter, . Oriented, spatial domain filters are constructed by imposing an appropriate angular selectivity in the frequency domain. A filter bank is designed, and outputs of the filters of the bank are steered to construct an oriented scale-space decomposition of an image subband. Some applications are discussed.