José Crespo

Theoretical aspects of morphological filters by reconstruction

Abstract This paper investigates morphological connected filters and, in particular, the so-called filters by reconstruction. A brief background is offered on the theory of morphological filtering. Then, the concept of connectivity is introduced within the morphological framework, which makes it possible to establish connected filters as those that do not introduce discontinuities or, in other words, that extend the input image flat zones. An important subset of connected filters is the class of filters by reconstruction, which allows to build connected filters that treat both the peaks and valleys of an input image while possessing a robustness property called the strong-property. The focus of our research is on the combination, by means of the sup- and inf-operations, of alternating filters by reconstruction when their component filters belong to a granulometry and an antigranulometry (by reconstruction). These operators will be investigated by means of the study of their grain and pore properties. Some commutation properties are introduced that facilitate the manipulation of filters by reconstruction. An important theoretical result of this paper is the establishment of a new family of strong morphological filters. Although most theoretical expressions refer to set operators, results are automatically extendable for non-binary (gray-level) functions.

Medical Data Analysis

In its first part, this contribution reviews shortly the application of neural network methods to medical problems and characterizes its advantages and problems in the context of the medical background. Successful application examples show that human diagnostic capabilities are significantly worse than the neural diagnostic systems. Then, paradigm of neural networks is shortly introduced and the main problems of medical data base and the basic approaches for training and testing a network by medical data are described. Additionally, the problem of interfacing the network and its result is given and the neurofuzzy approach is presented. Finally, as case study of neural rule based diagnosis septic shock diagnosis is described, on one hand by a growing neural network and on the other hand by a rule based system.

The flat zone approach: a general low-level region merging segmentation method

Abstract This paper presents a segmentation method, the flat zone approach, that avoids some limitations of the watershed-plus-markers method. The watershed-plus-markers approach, which is the traditional segmentation technique in mathematical morphology, has two inherent problems: (1) the possible separation of a piecewise-constant region of the input image into several regions in the output partition, and (2) the problem of obtaining markers (connected components of pixels signaling significant regions) for features that are one or two pixels wide. These problems are related to the limited resolution power (for feature extraction) of gradient operators. The flat zone approach extends the region marker concept (to contain the entire (and not part of) regions) and requires neither the computation of a gradient function nor the modification of the support of the input image in order to increase the size of the features. Our approach works on the graph formed by the image flat zones (or piecewise-constant regions). This fact ensures that the input image regions are not broken and can consider all input flat zones regardless their size. An inclusion relationship between the flat zones of the input image and the regions of the output partition is imposed. That is, a flat zone segmentation method is a region (flat zone) merging method and behaves like a connected operator. Our method is robust (in the sense that it is invariant under certain intensity value transformations) and uses a hierarchical waiting queue algorithm that makes it extremely efficient.

Locality and Adjacency Stability Constraints for MorphologicalConnected Operators

This paper investigates two constraints for the connected operator class. For binary images, connected operators are those that treat grains and pores of the input in an all or nothing way, and therefore they do not introduce discontinuities. The first constraint, called connected-component (c.c.) locality, constrains the part of the input that can be used for computing the output of each grain and pore. The second, called adjacency stability, establishes an adjacency constraint between connected components of the input set and those of the output set. Among increasing operators, usual morphological filters can satisfy both requirements. On the other hand, some (non-idempotent) morphological operators such as the median cannot have the adjacency stability property. When these two requirements are applied to connected and idempotent morphological operators, we are lead to a new approach to the class of filters by reconstruction. The important case of translation invariant operators and the relationships between translation invariance and connectivity are studied in detail. Concepts are developed within the binary (or set) framework; however, conclusions apply as well to flat non-binary (gray-level) operators.

On distributing load in cloud computing: A real application for very-large image datasets

Managing large image collections has become an important issue for information companies and institutions. We present a cloud computing service and its application for the storage and analysis of very-large images. This service has been implemented using multiple distributed and collaborative agents. For image storage and analysis, a regionoriented data structure is utilized, which allows storing and describing image regions using low-level descriptors. Different types of structural relationships between regions are also taken into account. A distinctive goal of this work is that data operations are adapted for working in a distributed mode. This allows that an input image can be divided into different sub-images that can be stored and processed separately by different agents in the system, facilitating processing very-large images in a parallel manner. A key aspect to decrease processing time for parallelized tasks is the use of an appropriate load balancer to distribute and assign tasks to agents with less workload.

OntoDataClean: ontology-based integration and preprocessing of distributed data

Within the knowledge discovery in databases (KDD) process, previous phases to data mining consume most of the time spent analysing data. Few research efforts have been carried out in theses steps compared to data mining, suggesting that new approaches and tools are needed to support the preparation of data. As regards, we present in this paper a new methodology of ontology-based KDD adopting a federated approach to database integration and retrieval. Within this model, an ontology-based system called OntoDataClean has been developed dealing with instance-level integration and data preprocessing. Within the OntoDataClean development, a preprocessing ontology was built to store the information about the required transformations. Various biomedical experiments were carried out, showing that data have been correctly transformed using the preprocessing ontology. Although OntoDataClean does not cover every possible data transformation, it suggests that ontologies are a suitable mechanism to improve quality in the various steps of KDD processes.

Medical informatics and bioinformatics: European efforts to facilitate synergy

Over the past decade there have been several attempts to rethink the basic strategies and scope of medical informatics. Meanwhile, bioinformatics has only recently experienced a similar debate about its scientific character. Both disciplines envision the development of novel diagnostic, therapeutic, and management tools, and products for patient care. A combination of the expertise of medical informatics in developing clinical applications and the focused principles that have guided bioinformatics could create a synergy between the two areas of application. Such interaction could have a great influence on future health research and the ultimate goal, namely continuity and individualization of health care. This article summarizes current activities related to facilitating synergy between medical informatics and bioinformatics, emphasizing activities in Europe while relating them to efforts in other parts of the world. The report provides examples of the analysis that European investigators are carrying out, aiming to propose new ideas for collaborations between medical informatics and bioinformatics researchers in a variety of areas.

New results on the theory of morphological filters by reconstruction

This paper treats the problem of establishing bounds for the morphological filter by reconstruction class. Morphological filters by reconstruction, which are composed of openings and closings by reconstruction, are useful filters for image processing because they do not introduce discontinuities. The main contributions of this paper are: (a) To establish when the combination of openings by reconstruction (or, respectively, of closings by reconstruction) is an opening by reconstruction (respectively a closing by reconstruction). (b) To establish, for any filter by reconstruction, upper and lower bounds that are, respectively, a closing by reconstruction and an opening by reconstruction. In addition, the paper investigates certain aspects of filters by reconstruction that possess a robustness property called strong property. Some dual and equivalent forms are introduced for a family of multi-level filters recently introduced. A significant side-result is to determine some instances of connected openings composed by openings and closings by reconstruction that are not openings by reconstruction (similarly for closings).

The Flat Zone Approach and Color Images

This paper treats the application of the flat zone approach for color images. For gray-level images, the flat zone approach was presented in (Crespo & Serra 1993) as a segmentation approach that imposes an inclusion relationship between the flat zones (or piecewise-constant regions) of the input image and the regions of the output partition. That is, a flat zone segmentation method behaves like a connected operator. Color images are formed by several (a priori) independent bands. In this paper we discuss how color images need a different treatment from that used for gray-level images. For gray-level images, the flat zone inclusion relationship preserves the shapes of the features that are observed in the input. On the other hand, for color images this desirable shape-preservation effect would not be obtained by forcing the inclusion relationship between the flat zones of each band and the regions of the output partition. A mask that contains the important regions of the color image, computed for each color band, is employed to restrict the flat zone inclusion relationship to those flat zones belonging to the mask. As in the gray-level case, the presented color segmentation method uses a hierarchical waiting queue algorithm that makes it computationally efficient.

International Efforts in Nanoinformatics Research Applied to Nanomedicine

BACKGROUND Nanomedicine and nanoinformatics are novel disciplines facing substantial challenges. Since nanomedicine involves complex and massive data analysis and management, a new discipline named nanoinformatics is now emerging to provide the vision and the informatics methods and tools needed for such purposes. Methods from biomedi-cal informatics may prove applicable with some adaptation despite nanomedicine involving different biophysical and biochemical characteristics of nanomaterials and corresponding differences in information complexity. OBJECTIVES We analyze recent initiatives and opportunities for research in nanomedicine and nanoinformatics as well as the previous experience of the authors, particularly in the context of a European project named ACTION-Grid. In this project the authors aimed to create a collaborative environment in biomedical and nanomedical research among countries in Europe, Western Balkans, Latin America, North Africa and the USA. METHODS We review and analyze the rationale and scientific issues behind the new fields of nanomedicine and nanoinformatics. Such a review is linked to actual research projects and achievements of the authors within their groups. RESULTS The work of the authors at the intersection between these two areas is presented. We also analyze several research initiatives that have recently emerged in the EU and USA context and highlight some ideas for future action at the international level. CONCLUSIONS Nanoinformatics aims to build new bridges between medicine, nanotechnology and informatics, allowing the application of computational methods in the nano-related areas. Opportunities for world-wide collaboration are already emerging and will be influential in advancing the field.