José Manuel Bravo

A New Supervised Method for Blood Vessel Segmentation in Retinal Images by Using Gray-Level and Moment Invariants-Based Features

This paper presents a new supervised method for blood vessel detection in digital retinal images. This method uses a neural network (NN) scheme for pixel classification and computes a 7-D vector composed of gray-level and moment invariants-based features for pixel representation. The method was evaluated on the publicly available DRIVE and STARE databases, widely used for this purpose, since they contain retinal images where the vascular structure has been precisely marked by experts. Method performance on both sets of test images is better than other existing solutions in literature. The method proves especially accurate for vessel detection in STARE images. Its application to this database (even when the NN was trained on the DRIVE database) outperforms all analyzed segmentation approaches. Its effectiveness and robustness with different image conditions, together with its simplicity and fast implementation, make this blood vessel segmentation proposal suitable for retinal image computer analyses such as automated screening for early diabetic retinopathy detection.

Brief Guaranteed state estimation by zonotopes

This paper presents a new approach to guaranteed state estimation for non-linear discrete-time systems with a bounded description of noise and parameters. The main result is an algorithm to compute a set that contains the states consistent with the measured output and the given noise and parameters. This set is represented by a zonotope. The size of the zonotope is minimized each sample time by an analytic expression or by solving a convex optimization problem. Interval arithmetic is used to calculate a guaranteed trajectory of the process state. Two examples have been provided to clarify the algorithm.

Input-to-State Stability: A Unifying Framework for Robust Model Predictive Control

This paper deals with the robustness of Model Predictive Controllers for constrained uncertain nonlinear systems. The uncertainty is assumed to be modeled by a state and input dependent signal and a disturbance signal. The framework used for the analysis of the robust stability of the systems controlled by MPC is the wellknown Input-to-State Stability. It is shown how this notion is suitable in spite of the presence of constraints on the system and of the possible discontinuity of the control law.

Guaranteed state estimation by zonotopes

This paper presents a new approach to guaranteed state estimation for nonlinear discrete-time systems with a bounded description of noise and parameters. The main result is an algorithm to compute a set that contains the states consistent with the measured output and the given noise and parameters. This set is represented by a zonotope. The volume of the zonotope is minimized each sample instant solving a convex optimization problem. Interval arithmetic is used to calculate a guaranteed trajectory of the state process. Two examples have been provided for clarifying the algorithm.

Bounded error identification of systems with time-varying parameters

This paper presents a new approach to guaranteed system identification for time-varying parameterized discrete-time systems. A bounded description of noise in the measurement is considered. The main result is an algorithm to compute a set that contains the parameters consistent with the measured output and the given bound of the noise. This set is represented by a zonotope, that is, an affine map of a unitary hypercube. A recursive procedure minimizes the size of the zonotope with each noise corrupted measurement. The zonotope allows us to take into account the time-varying nature of the parameters in a non conservative way. An example has been provided to clarify the algorithm.

Robust MPC of constrained discrete-time nonlinear systems based on approximated reachable sets

A robust MPC for constrained nonlinear systems with uncertainties is presented. Outer bounds of the reachable sets of the system are used to predict the evolution of the system under uncertainty. A method that uses zonotopes to represent the approximated reachable sets is proposed. The closed-loop system is ultimately bounded thanks to a contractive constraint that drives the system to a robust invariant set.

Brief paper: A set-membership state estimation algorithm based on DC programming

This paper presents a new approach to guaranteed state estimation for nonlinear discrete-time systems with a bounded description of noise and parameters. The sets of states that are consistent with the evolution of the system, the measured outputs and bounded noise and parameters are represented by zonotopes. DC programming and intersection operations are used to obtain a tight bound. An example is given to illustrate the proposed algorithm.

Obtaining optic disc center and pixel region by automatic thresholding methods on morphologically processed fundus images

Development of automatic retinal disease diagnosis systems based on retinal image computer analysis can provide remarkably quicker screening programs for early detection. Such systems are mainly focused on the detection of the earliest ophthalmic signs of illness and require previous identification of fundal landmark features such as optic disc (OD), fovea or blood vessels. A methodology for accurate center-position location and OD retinal region segmentation on digital fundus images is presented in this paper. The methodology performs a set of iterative opening-closing morphological operations on the original retinography intensity channel to produce a bright region-enhanced image. Taking blood vessel confluence at the OD into account, a 2-step automatic thresholding procedure is then applied to obtain a reduced region of interest, where the center and the OD pixel region are finally obtained by performing the circular Hough transform on a set of OD boundary candidates generated through the application of the Prewitt edge detector. The methodology was evaluated on 1200 and 1748 fundus images from the publicly available MESSIDOR and MESSIDOR-2 databases, acquired from diabetic patients and thus being clinical cases of interest within the framework of automated diagnosis of retinal diseases associated to diabetes mellitus. This methodology proved highly accurate in OD-center location: average Euclidean distance between the methodology-provided and actual OD-center position was 6.08, 9.22 and 9.72 pixels for retinas of 910, 1380 and 1455 pixels in size, respectively. On the other hand, OD segmentation evaluation was performed in terms of Jaccard and Dice coefficients, as well as the mean average distance between estimated and actual OD boundaries. Comparison with the results reported by other reviewed OD segmentation methodologies shows our proposal renders better overall performance. Its effectiveness and robustness make this proposed automated OD location and segmentation method a suitable tool to be integrated into a complete prescreening system for early diagnosis of retinal diseases.

Stability analysis and synthesis of multivariable fuzzy systems using interval arithmetic

Abstract This paper deals with the design of stable rule-based fuzzy control systems. Interval analysis is applied to design a stable fuzzy Takagi–Sugeno–Kang controller using a robust condition to ensure the stability. The presented methodology starts with a state model of the plant, finds a candidate fuzzy controller and uses an interval arithmetic algorithm to verify the stability of closed-loop fuzzy model. It is important to emphasize the generality of the presented methodology for fuzzy controller synthesis since there are no constraints in the state vector nor in the control vector. This methodology can also be used with nonlinear plant models. In previous works we showed the applicability of the interval analysis to design a controller that ensures the stability of first order nonlinear system. In this paper, we extend the analysis and the synthesis of stable fuzzy control system to the multivariable case. An example with a fuzzy controller for a nonlinear system is presented to illustrate the design procedure.

Locating the fovea center position in digital fundus images using thresholding and feature extraction techniques.

A new methodology for detecting the fovea center position in digital retinal images is presented in this paper. A pixel is firstly searched for within the foveal region according to its known anatomical position relative to the optic disc and vascular tree. Then, this pixel is used to extract a fovea-containing subimage on which thresholding and feature extraction techniques are applied so as to find fovea center. The methodology was evaluated on 1200 fundus images from the publicly available MESSIDOR database, 660 of which present signs of diabetic retinopathy. In 93.92% of these images, the distance between the methodology-provided and actual fovea center position remained below 1/4 of one standard optic disc radius (i.e., 17, 26, and 27 pixels for MESSIDOR retinas of 910, 1380 and 1455 pixels in size, respectively). These results outperform all the reviewed methodologies available in literature. Its effectiveness and robustness with different illness conditions makes this proposal suitable for retinal image computer analyses such as automated screening for early diabetic retinopathy detection.