Patricio García Báez

Artificial Intelligent Systems Based on Supervised HUMANN for Differential Diagnosis of Cognitive Impairment: Towards a 4P-HCDS

Differential and early diagnosis of cognitive impairment (CI) continues being one of the crucial points to which clinical medicine faces at every level of attention, and a significant public health concern. This work proposes new CI diagnostic tools based on a data fusion scheme, artificial neural networks and ensemble systems. Concretely we have designed a supervised HUMANN [1] with capacity of missing data processing (HUMANN-S) and a HUMANN-S ensemble system. These intelligent diagnostic systems are inside EDEVITALZH, a clinical virtual environment to assist the diagnosis and prognosis of CI, Alzheimers disease and other dementias. Our proposal is a personalized, predictive, preventive, and participatory-healthcare delivery system (4P-HCDS) and is an optimal solution for an e-health framework. We explore their ability presenting preliminary results on differential diagnosis of CI using neuropsychological tests from 267 consultations on 30 patients by the Alzheimers Patient Association of Gran Canaria.

A parametric study of HUMANN in relation to the noise: application for the identification of compounds of environmental interest

In this paper we present a parametric study of a hierarchical unsupervised modular adaptive neural network (HUMANN), in dealing with noise. HUMANN is a biologically plausible feedforward neural architecture which has the capacity for working in domains with noise and overlapping classes, with no priori information of the number of different classes in the data, with highly non-linear boundary class and with high dimensionality data vectors. It is appropriate for classification processes performing blind clustering. The study has been accomplished round the two most noise-dependent HUMANN parameters, λ and ρ, using synthesized databases (sinusoidal signals with Gaussian noise). We show that HUMANN is highly resistant to noise, improving the performance of different neural architectures such as ART2 and DIGNET. We also present the application of HUMANN for the identification of pollutants in the environment. Specifically it has been tested with Polychlorinated dibenzofurans (PCDFs), some of the most hazardous pollutants of the environment.

Automatic prognostic determination and evolution of cognitive decline using artificial neural networks

This work tries to go a step further in the development of methods based on automatic learning techniques to parse and interpret data relating to cognitive decline (CD). There have been studied the neuropsychological tests of 267 consultations made over 30 patients by the Alzheimers Patient Association of Gran Canaria in 2005. The Sanger neural network adaptation for missing values treatment has allowed making a Principal Components Analysis (PCA) on the successfully obtained data. The results show that the first three obtained principal components are able to extract information relating to functional, cognitive and instrumental sintomatology, respectively, from the test. By means of these techniques, it is possible to develop tools that allow physicians to quantify, view and make a better pursuit of the sintomatology associated to the cognitive decline processes, contributing to a better knowledge of these ones.

HUMANN-based system to identify benzimidazole fungicides using multi-synchronous fluorescence spectra: An ensemble approach

AbstractIn this paper, we approach, using neural computation and ensemble systems, a pattern classification problem in fluorescence spectrometry, the resolution of difficult multi-fungicide mixtures (overlapping), specifically the benzimidazole fungicides, benomyl, carbendazim, thiabendazole and fuberidazole. These fungicides are compounds of an important environmental interest. Because of this, from an analytical point of view, it is interesting to develop sensitive, selective and simple methods for their determination. Fluorescence spectrometry has proven to be a sensitive and selective technique for determination of many compounds of environmental interest, but in some cases it is not enough. HUMANN is a hierarchical, unsupervised, modular, adaptive neural net with high biological plausibility, which has shown to be suitable for identification of these fungicides and organochlorinated compounds of environmental interest. We propose two modular artificial intelligent systems, with a structure of pre-processing and processing stage, a multi-input HUMANN-based system, using multi-fluorescence spectra as input to the system, and a HUMANN-ensemble system. We analyze the optimal configuration of inputs and the ensemble in order to obtain better results. We study such figures as precision and sensitivity of the method. Our proposal is a smart, flexible and effective complementary method, which allows reducing the analytical and/or computational complexity of the analysis. FigureStages in identification of benzimidazole fungicides

GaNEn: A new gating neural ensemble for automatic assessment of the Severity Level of Dementia using neuropsychological tests

Dementia is one of the associated diseases to aging most prevalents. An important issue about this neuropathology, as of yet unsolved, is the absence of therapeutic tools that manage or stop its progression and symptoms in a constant and supported way. In the present study, we propose a new computational intelligent tool to diagnose the Severity Level of Dementia (SLD) using gating neural network and neural ensemble approaches. We present a gating neural ensemble (GaNEn). This system is a new formulation of a neural network ensemble, where the gating neural network takes part in the combination strategy of ensemble system, and the main expert module in its construction is a HUMANN-S (Supervised HUMANN (Hybrid Unsupervised Modular Artificial Neural Network)) architecture. GaNEn is characterized by an incremental capacity concerning missing data management and their influence in the final diagnosis. It improves previous computational solutions and obtains higher accuracy diagnosis. The GaNEn system is a significant achievement in the medical diagnosis of neurological disorders because it could aid in the design of pharmaco-therapeutic strategies to contain dementia. It is also capable of supplying the best neuropsychological scales for dementia severity grades. We have explored its ability using a battery of neuropsychological tests from people with Alzheimer type dementia (AD), Vascular type dementia (VD) and other dementia type (OD) like Trauma, Subcortical, Parkinson and Infectious, from the Alzheimers Association of Gran Canaria.

An Ensemble Approach for the Diagnosis of Cognitive Decline with Missing Data

This work applies new techniques of automatic learning to diagnose neuro decline processes usually related to aging. Early detection of cognitive decline (CD) is an advisable practice under multiple perspectives. A study of neuropsychological tests from 267 consultations on 30 patients by the Alzheimers Patient Association of Gran Canaria is carried out. We designed neural computational CD diagnosis systems, using a multi-net and ensemble structure that is applied to the treatment of missing data present in consultations. The results show significant improvements over simple classifiers. These systems would allow applying policies of early detection of dementias in primary care centers where specialized professionals are not present.

Towards a Model of Volume Transmission in Biological and Artificial Neural Networks: A CAST Approach

At present, a new type of process for signalling between cells seems to be emerging, the diffusion or volume transmission. The volume transmission is performed by means of a gas diffusion process, which is obtained with a diffusive type of signal (NO). We present in this paper a CAST approach, in order to develop a NOdi ffusion model, away from a biologically plausible morphology, that provides a formal framework for the establishing of neural signalling capacity of NOin biological and artificial neural environments. It is also presented a study which shows implications of volume transmission in the emergence of complex structures and self-organisation processes in both biological and artificial neural netwoks. Finally, we present the diffusion version of the Associative Network (AN) [6], the Diffusion Associative Network (DAN), where a more general framework of neural learning, which is based in synaptic and volume transmission, is considered.

Self-Organizing Maps for Early Detection of Denial of Service Attacks

Detection and early alert of Denial of Service (DoS) attacks are very important actions to make appropriate decisions in order to minimize their negative impact. DoS attacks have been catalogued as of high-catastrophic index and hard to defend against. Our study presents advances in the area of computer security against DoS attacks. In this chapter, a flexible method is presented, capable of effectively tackling and overcoming the challenge of DoS (and distributed DoS) attacks using a CISDAD (Computer Intelligent System for DoS Attacks Detection). It is a hybrid intelligent system with a modular structure: a pre-processing module (non neural) and a processing module based on Kohonen Self-Organizing artificial neural networks. The proposed system introduces an automatic differential detection of several Normal Traffic and several Toxic Traffics, clustering them upon its Transport-Layer-Protocol behavior. Two computational studies of CISDAD working with real networking traffic will be described, showing a high level of effectiveness in the CISDAD detection process. Finally, in this chapter, the possibility for specific adaptation to the Healthcare environment that CISDAD can offer is introduced.

Neural Computation Methods in the Determination of Fungicides

Fungicides are a specific type of pesticides that control diseases produced by fungi, inhibiting specifically or directly killing these parasite organisms. It has been used during more than forty years for the protection of harvests and farming lands. There are several treatments with fungicides: protector or preventative from the germination of the spores and follow up infection, or curing or eradicating, when mycelium has been formed and must be controlled. Treatments are applied in the soil and on stored vegetable products, seeds or plants. In this case two types of fungicides are classified: a) those with contact: unable to penetrate the inside of the vegetable and control epiphytic fungi, and b) those that are systemic and control endophyte fungi. The large variety of fungicides that exist makes a classification absolutely necessary. We find different taxonomies that can be studied not only for their structural aspects, but such as chemical composition, and also in action modes. The guidelines that regulate the managed use and classification of these are defined in Spanish, European and American Environmental Agencies. The use of chemical products in farming activities has produced important benefits in agricultural production, increasing profitability of harvests while simultaneously raising the quality levels of the food products. Nevertheless, there are other considerations with regard to these benefits that result in the systematic destruction of parasites, that affect the health of the plants, animals and human health, and require consideration of the interaction of the different chemical main components with animal species and with humans themselves (Rivas, 2004). To begin with, the form of the administracion of fungicides favors their accumulation in the sediments and in drainage waters. Also, in cases where plaguicides have been used indiscriminately, the species of plagues have become resistent and difficult to control. The main source of exposition of the general population to fungicides is through food, a fact that has forced the establishment of regulations of its maximum daily ingestion allowance. There are studies that relate the exposure to pesticides with damaging effects on human health: neurological damage, hormonal and reproductive disorders, dermatological or carcinogenic reactions (Alavanja, 2004).

Extension of HUMANN for Dealing with Noise and with Classes of Different Shape and Size: A Parametric Study

In this paper an extension of HUMANN (hierarchical unsupervised modular adaptive neural network) is presented together with a parametric study of this network in dealing with noise and with classes of any shape and size. The study has been made based on the two most noise dependent HUMANN parameters, [and], using synthesised databases (bidimensional patterns with outliers and classes with different probability density distribution). In order to evaluate the robustness of HUMANN a Monte Carlo [1] analysis was carried out using the creation of separate data in given classes. The influence of the different parameters in the recovery of these classes was then studied.