Network-based stochastic semisupervised learning

Semisupervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In this paper, we propose a semisupervised data classification model based on a combined random-preferential walk of particles in a network (graph) constructed from the input dataset. The particles of the same class cooperate among themselves, while the particles of different classes compete with each other to propagate class labels to the whole network. A rigorous model definition is provided via a nonlinear stochastic dynamical system and a mathematical analysis of its behavior is carried out. A numerical validation presented in this paper confirms the theoretical predictions. An interesting feature brought by the competitive-cooperative mechanism is that the proposed model can achieve good classification rates while exhibiting low computational complexity order in comparison to other network-based semisupervised algorithms. Computer simulations conducted on synthetic and real-world datasets reveal the effectiveness of the model.

A New Prediction Model for Slope Stability Analysis

The instability of river bank can result in considerable human and land losses. The Po river is the most important in Italy, characterized by main banks of significant and constantly increasing height. This study presents multilayer perceptron of artificial neural network (ANN) to construct prediction models for the stability analysis of river banks along the Po River, under various river and groundwater boundary conditions. For this aim, a number of networks of threshold logic unit are tested using different combinations of the input parameters. Factor of safety (FS), as an index of slope stability, is formulated in terms of several influencing geometrical and geotechnical parameters. In order to obtain a comprehensive geotechnical database, several cone penetration tests from the study site have been interpreted. The proposed models are developed upon stability analyses using finite element code over different representative sections of river embankments. For the validity verification, the ANN models are employed to predict the FS values of a part of the database beyond the calibration data domain. The results indicate that the proposed ANN models are effective tools for evaluating the slope stability. The ANN models notably outperform the derived multiple linear regression models.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

An insulin infusion advisory system based on autotuning nonlinear model-predictive control

This paper aims at the development and evaluation of a personalized insulin infusion advisory system (IIAS), able to provide real-time estimations of the appropriate insulin infusion rate for type 1 diabetes mellitus (T1DM) patients using continuous glucose monitors and insulin pumps. The system is based on a nonlinear model-predictive controller (NMPC) that uses a personalized glucose-insulin metabolism model, consisting of two compartmental models and a recurrent neural network. The model takes as input patient's information regarding meal intake, glucose measurements, and insulin infusion rates, and provides glucose predictions. The predictions are fed to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. An algorithm based on fuzzy logic has been developed for the on-line adaptation of the NMPC control parameters. The IIAS has been in silico evaluated using an appropriate simulation environment (UVa T1DM simulator). The IIAS was able to handle various meal profiles, fasting conditions, interpatient variability, intraday variation in physiological parameters, and errors in meal amount estimations.

An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.

Visuospatial working memory in very preterm and term born children - Impact of age and performance

Working memory is crucial for meeting the challenges of daily life and performing academic tasks, such as reading or arithmetic. Very preterm born children are at risk of low working memory capacity. The aim of this study was to examine the visuospatial working memory network of school-aged preterm children and to determine the effect of age and performance on the neural working memory network. Working memory was assessed in 41 very preterm born children and 36 term born controls (aged 7–12 years) using functional magnetic resonance imaging (fMRI) and neuropsychological assessment. While preterm children and controls showed equal working memory performance, preterm children showed less involvement of the right middle frontal gyrus, but higher fMRI activation in superior frontal regions than controls. The younger and low-performing preterm children presented an atypical working memory network whereas the older high-performing preterm children recruited a working memory network similar to the controls. Results suggest that younger and low-performing preterm children show signs of less neural efficiency in frontal brain areas. With increasing age and performance, compensational mechanisms seem to occur, so that in preterm children, the typical visuospatial working memory network is established by the age of 12 years.

Multi-Model Data Fusion to Improve an Early Warning System for Hypo-/Hyperglycemic Events

Correct predictions of future blood glucose levels in individuals with Type 1 Diabetes (T1D) can be used to provide early warning of upcoming hypo-/hyperglycemic events and thus to improve the patient's safety. To increase prediction accuracy and efficiency, various approaches have been proposed which combine multiple predictors to produce superior results compared to single predictors. Three methods for model fusion are presented and comparatively assessed. Data from 23 T1D subjects under sensor-augmented pump (SAP) therapy were used in two adaptive data-driven models (an autoregressive model with output correction - cARX, and a recurrent neural network - RNN). Data fusion techniques based on i) Dempster-Shafer Evidential Theory (DST), ii) Genetic Algorithms (GA), and iii) Genetic Programming (GP) were used to merge the complimentary performances of the prediction models. The fused output is used in a warning algorithm to issue alarms of upcoming hypo-/hyperglycemic events. The fusion schemes showed improved performance with lower root mean square errors, lower time lags, and higher correlation. In the warning algorithm, median daily false alarms (DFA) of 0.25%, and 100% correct alarms (CA) were obtained for both event types. The detection times (DT) before occurrence of events were 13.0 and 12.1 min respectively for hypo-/hyperglycemic events. Compared to the cARX and RNN models, and a linear fusion of the two, the proposed fusion schemes represents a significant improvement.

Image sub-segmentation by PFCM and Artificial Neural Networks to detect pore space in 2D and 3D CT soil images

The image by Computed Tomography is a non-invasive alternative for observing soil structures, mainly pore space. The pore space correspond in soil data to empty or free space in the sense that no material is present there but only fluids, the fluid transport depend of pore spaces in soil, for this reason is important identify the regions that correspond to pore zones. In this paper we present a methodology in order to detect pore space and solid soil based on the synergy of the image processing, pattern recognition and artificial intelligence. The mathematical morphology is an image processing technique used for the purpose of image enhancement. In order to find pixels groups with a similar gray level intensity, or more or less homogeneous groups, a novel image sub-segmentation based on a Possibilistic Fuzzy c-Means (PFCM) clustering algorithm was used. The Artificial Neural Networks (ANNs) are very efficient for demanding large scale and generic pattern recognition applications for this reason finally a classifier based on artificial neural network is applied in order to classify soil images in two classes, pore space and solid soil respectively.

Using modular neural networks to model self-consciousness and self-representation for artificial entities

Self-consciousness implies not only self or group recognition, but also real knowledge of one’s own identity. Self-consciousness is only possible if an individual is intelligent enough to formulate an abstract self-representation. Moreover, it necessarily entails the capability of referencing and using this elf-representation in connection with other cognitive features, such as inference, and the anticipation of the consequences of both one’s own and other individuals’ acts. In this paper, a cognitive architecture for self-consciousness is proposed. This cognitive architecture includes several modules: abstraction, self-representation, other individuals'representation, decision and action modules. It includes a learning process of self-representation by direct (self-experience based) and observational learning (based on the observation of other individuals). For model implementation a new approach is taken using Modular Artificial Neural Networks (MANN). For model testing, a virtual environment has been implemented. This virtual environment can be described as a holonic system or holarchy, meaning that it is composed of autonomous entities that behave both as a whole and as part of a greater whole. The system is composed of a certain number of holons interacting. These holons are equipped with cognitive features, such as sensory perception, and a simplified model of personality and self-representation. We explain holons’ cognitive architecture that enables dynamic self-representation. We analyse the effect of holon interaction, focusing on the evolution of the holon’s abstract self-representation. Finally, the results are explained and analysed and conclusions drawn.

The Optimal combination: Grammatical Swarm, Particle Swarm Optimization and Neural Networks.

Social behaviour is mainly based on swarm colonies, in which each individual shares its knowledge about the environment with other individuals to get optimal solutions. Such co-operative model differs from competitive models in the way that individuals die and are born by combining information of alive ones. This paper presents the particle swarm optimization with differential evolution algorithm in order to train a neural network instead the classic back propagation algorithm. The performance of a neural network for particular problems is critically dependant on the choice of the processing elements, the net architecture and the learning algorithm. This work is focused in the development of methods for the evolutionary design of artificial neural networks. This paper focuses in optimizing the topology and structure of connectivity for these networks.

Polynomial approximation using particle swarm optimization of lineal enhanced neural networks with no hidden layers.

This paper presents some ideas about a new neural network architecture that can be compared to a Taylor analysis when dealing with patterns. Such architecture is based on lineal activation functions with an axo-axonic architecture. A biological axo-axonic connection between two neurons is defined as the weight in a connection in given by the output of another third neuron. This idea can be implemented in the so called Enhanced Neural Networks in which two Multilayer Perceptrons are used; the first one will output the weights that the second MLP uses to computed the desired output. This kind of neural network has universal approximation properties even with lineal activation functions. There exists a clear difference between cooperative and competitive strategies. The former ones are based on the swarm colonies, in which all individuals share its knowledge about the goal in order to pass such information to other individuals to get optimum solution. The latter ones are based on genetic models, that is, individuals can die and new individuals are created combining information of alive one; or are based on molecular/celular behaviour passing information from one structure to another. A swarm-based model is applied to obtain the Neural Network, training the net with a Particle Swarm algorithm.

Artificial metaplasticity prediction model for cognitive rehabilitation outcome in acquired brain injury patients

Objective The main purpose of this research is the novel use of artificial metaplasticity on multilayer perceptron (AMMLP) as a data mining tool for prediction the outcome of patients with acquired brain injury (ABI) after cognitive rehabilitation. The final goal aims at increasing knowledge in the field of rehabilitation theory based on cognitive affectation. Methods and materials The data set used in this study contains records belonging to 123 ABI patients with moderate to severe cognitive affectation (according to Glasgow Coma Scale) that underwent rehabilitation at Institut Guttmann Neurorehabilitation Hospital (IG) using the tele-rehabilitation platform PREVIRNEC©. The variables included in the analysis comprise the neuropsychological initial evaluation of the patient (cognitive affectation profile), the results of the rehabilitation tasks performed by the patient in PREVIRNEC© and the outcome of the patient after a 3–5 months treatment. To achieve the treatment outcome prediction, we apply and compare three different data mining techniques: the AMMLP model, a backpropagation neural network (BPNN) and a C4.5 decision tree. Results The prediction performance of the models was measured by ten-fold cross validation and several architectures were tested. The results obtained by the AMMLP model are clearly superior, with an average predictive performance of 91.56%. BPNN and C4.5 models have a prediction average accuracy of 80.18% and 89.91% respectively. The best single AMMLP model provided a specificity of 92.38%, a sensitivity of 91.76% and a prediction accuracy of 92.07%. Conclusions The proposed prediction model presented in this study allows to increase the knowledge about the contributing factors of an ABI patient recovery and to estimate treatment efficacy in individual patients. The ability to predict treatment outcomes may provide new insights toward improving effectiveness and creating personalized therapeutic interventions based on clinical evidence.

Data Mining with Enhanced Neural Networks-CMMSE

Abstract This paper presents a new method to extract knowledge from existing data sets, that is, to extract symbolic rules using the weights of an Artificial Neural Network. The method has been applied to a neural network with special architecture named Enhanced Neural Network (ENN). This architecture improves the results that have been obtained with multilayer perceptron (MLP). The relationship among the knowledge stored in the weights, the performance of the network and the new implemented algorithm to acquire rules from the weights is explained. The method itself gives a model to follow in the knowledge acquisition with ENN.

Dam seepage analysis based on artificial neural networks: the hysteresis phenomenon

Seepage flow measurement is an important behavior indicator when providing information about dam performance. The main objective of this study is to analyze seepage by means of an artificial neural network model. The model is trained and validated with data measured at a case study. The dam behavior towards different water level changes is reproduced by the model and a hysteresis phenomenon detected and studied. Artificial neural network models are shown to be a powerful tool for predicting and understanding seepage phenomenon.

Monitorización y optimización de terapias de tele-neurorrehabilitación cognitiva

El Daño Cerebral Adquirido (DCA) se define como una lesión cerebral que ocurre después del nacimiento y que no guarda relación con defectos congénitos o enfermedades degenerativas. En el cerebro, se llevan a cabo las funciones mentales superiores como la atención, la memoria, las funciones ejecutivas y el lenguaje, consideradas pre-requisitos básicos de la inteligencia. Sea cual sea su causa, todo daño cerebral puede afectar a una o varias de estas funciones, de ahí la gravedad del problema. A pesar de los avances en nuevas técnicas de intervención precoz y el desarrollo de los cuidados intensivos, las afectaciones cerebrales aún no tienen tratamiento ni quirúrgico ni farmacológico que permita una restitución de las funciones perdidas. Los tratamientos de neurorrehabilitación cognitiva y funcional pretenden, por tanto, la minimización o compensación de las alteraciones ocasionadas por una lesión en el sistema nervioso. En concreto, la rehabilitación cognitiva se define como el proceso en el que personas que han sufrido un daño cerebral trabajan de manera conjunta con profesionales de la salud para remediar o aliviar los déficits cognitivos surgidos como consecuencia de un episodio neurológico. Esto se consigue gracias a la naturaleza plástica del sistema nervioso, donde el cerebro es capaz de reconfigurar sus conexiones neuronales, tanto creando nuevas como modificando las ya existentes. Durante los últimos años hemos visto una transformación de la sociedad, en lo que se ha denominado "sociedad de la información", cuyo pilar básico son las Tecnologías de la Información y las Comunicaciones (TIC). La aplicación de estas tecnologías en medicina ha revolucionado la manera en que se proveen los servicios sanitarios. Así, donde tecnología y medicina se mezclan, la telerrehabilitación se define como la rehabilitación a distancia, ayudando a extender los servicios de rehabilitación más allá de los centros hospitalarios, rompiendo las barreras geográficas, mejorando la eficiencia de los procesos y monitorizando en todo momento el estado y evolución del paciente. En este contexto, el objetivo general de la presente tesis es mejorar la rehabilitación neuropsicológica de pacientes que sufren alteraciones cognitivas, mediante el diseño, desarrollo y validación de un sistema de telemedicina que incorpora las TIC para avanzar hacia un nuevo paradigma personalizado, ubicuo y ecológico. Para conseguirlo, se han definido los siguientes objetivos específicos: • Analizar y modelar un sistema de telerrehabilitación, mediante la definición de objetivos y requisitos de usuario para diseñar las diferentes funcionalidades necesarias. • Definir una arquitectura de telerrehabilitación escalable para la prestación de diferentes servicios que agrupe las funcionalidades necesarias en módulos. • Diseñar y desarrollar la plataforma de telerrehabilitación, incluida la interfaz de usuario, creando diferentes roles de usuario con sus propias funcionalidades. • Desarrollar de un módulo de análisis de datos para extraer conocimiento basado en los resultados históricos de las sesiones de rehabilitación almacenadas en el sistema. • Evaluación de los resultados obtenidos por los pacientes después del programa de rehabilitación, obteniendo conclusiones sobre los beneficios del servicio implementado. • Evaluación técnica de la plataforma de telerrehabilitación, así como su usabilidad y la relación coste/beneficio. • Integración de un dispositivo de eye-tracking que permita la monitorización de la atención visual mientras los pacientes ejecutan tareas de neurorrehabilitación. •Diseño y desarrollo de un entorno de monitorización que permita obtener patrones de atención visual. Como resumen de los resultados obtenidos, se ha desarrollado y validado técnicamente la plataforma de telerrehabilitación cognitiva, demostrando la mejora en la eficiencia de los procesos, sin que esto resulte en una reducción de la eficacia del tratamiento. Además, se ha llevado a cabo una evaluación de la usabilidad del sistema, con muy buenos resultados. Respecto al módulo de análisis de datos, se ha diseñado y desarrollado un algoritmo que configura y planifica sesiones de rehabilitación para los pacientes, de manera automática, teniendo en cuenta las características específicas de cada paciente. Este algoritmo se ha denominado Intelligent Therapy Assistant (ITA). Los resultados obtenidos por el asistente muestran una mejora tanto en la eficiencia como en la eficacia de los procesos, comparado los resultados obtenidos con los de la planificación manual llevada a cabo por los terapeutas. Por último, se ha integrado con éxito el dispositivo de eye-tracking en la plataforma de telerrehabilitación, llevando a cabo una prueba con pacientes y sujetos control que ha demostrado la viabilidad técnica de la solución, así como la existencia de diferencias en los patrones de atención visual en pacientes con daño cerebral. ABSTRACT Acquired Brain Injury (ABI) is defined as brain damage that suddenly and unexpectedly appears in people’s life, being the main cause of disability in developed countries. The brain is responsible of the higher cognitive functions such as attention, memory, executive functions or language, which are considered basic requirements of the intelligence. Whatever its cause is, every ABI may affects one or several functions, highlighting the severity of the problem. New techniques of early intervention and the development of intensive ABI care have noticeably improved the survival rate. However, despite these advances, brain injuries still have no surgical or pharmacological treatment to re-establish lost functions. Cognitive rehabilitation is defined as a process whereby people with brain injury work together with health service professionals and others to remediate or alleviate cognitive deficits arising from a neurological insult. This is achieved by taking advantage of the plastic nature of the nervous system, where the brain can reconfigure its connections, both creating new ones, and modifying the previously existing. Neuro-rehabilitation aims to optimize the plastic nature by inducing a reorganization of the neural network, based on specific experiences. Personalized interventions from individual impairment profile will be necessary to optimize the remaining resources by potentiating adaptive responses and inhibiting maladaptive changes. In the last years, some applications and software programs have been developed to train or stimulate cognitive functions of different neuropsychological disorders, such as ABI, Alzheimer, psychiatric disorders, attention deficit or hyperactivity disorder (ADHD). The application of technologies into medicine has changed the paradigm. Telemedicine allows improving the quality of clinical services, providing better access to them and helping to break geographical barriers. Moreover, one of the main advantages of telemedicine is the possibility to extend the therapeutic processes beyond the hospital (e.g. patient's home). As a consequence, a reduction of unnecessary costs and a better costs/benefits ratio are achieved, making possible a more efficient use of the available resources In this context, the main objective of this work is to improve neuro-rehabilitation of patients suffering cognitive deficits, by designing, developing and validating a telemedicine system that incorporates ICTs to change this paradigm, making it more personalized, ubiquitous and ecologic. The following specific objectives have been defined: • To analyse and model a tele-rehabilitation system, defining objectives and user requirements to design the different needed functionalities. • To define a scalable tele-rehabilitation architecture to offer different services grouping functionalities into modules. • To design and develop the tele-rehabilitation platform, including the graphic user interface, creating different user roles and permissions. • To develop a data analysis module to extract knowledge based on the historic results from the rehabilitation sessions stored in the system. • To evaluate the obtained results by patients after the rehabilitation program, arising conclusions about the benefits of the implemented service. • To technically evaluate the tele-rehabilitation platform, and its usability and the costs/benefit ratio. • To integrate an eye-tracking device allowing the monitoring of the visual attention while patients execute rehabilitation tasks. •To design and develop a monitoring environment that allows to obtain visual attention patterns. Summarizing the obtained results, the cognitive tele-rehabilitation platform has been developed and evaluated technically, demonstrating the improvements on the efficiency without worsening the efficacy of the process. Besides, a usability evaluation has been carried out, with very good results. Regarding the data analysis module, an algorithm has been designed and developed to automatically select and configure rehabilitation sessions, taking into account the specific characteristics of each patient. This algorithm is called Intelligent Therapy Assistant (ITA). The obtained results show an improvement both in the efficiency and the efficacy of the process, comparing the results obtained by patients when they receive treatments scheduled manually by therapists. Finally, an eye-tracking device has been integrated in the tele-rehabilitation platform, carrying out a study with patients and control subjects demonstrating the technical viability of the developed monitoring environment. First results also show that there are differences between the visual attention patterns between ABI patients and control subjects.