Elsa Ferreira Gomes

Classifying heart sounds using multiresolution time series motifs: an exploratory study

The aim of this work is to describe an exploratory study on the use of a SAX-based Multiresolution Motif Discovery method for Heart Sound Classification. The idea of our work is to discover relevant frequent motifs in the audio signals and use the discovered motifs and their frequency as characterizing attributes. We also describe different configurations of motif discovery for defining attributes and compare the use of a decision tree based algorithm with random forests on this kind of data. Experiments were performed with a dataset obtained from a clinic trial in hospitals using the digital stethoscope DigiScope. This exploratory study suggests that motifs contain valuable information that can be further exploited for Heart Sound Classification.

Numerical modelling of a gravity settler in dynamic conditions

An iterative numerical technique has been developed to simulate in detail the dynamics of a shallow-layer gravity settler. Currently acknowledged models apply only to specific equipments at steady-state and laboratory scale [Jeelani SAK, Hartland S. The continuous separation of liquid/liquid dispersions. Chem Eng Sci 1993;48(2):239-54]. To our knowledge, no study has ever addressed the dynamic simulation of a gravity settler. In this paper a direct numerical technique is presented for computing the thickness and drop-size composition of the dispersion band formed in a shallow-layer settler under steady-state and transient conditions. This technique is an extension for the settler of the one used on the stirred vessel by Ribeiro [Ribeiro LM. Simulacao Dinamica de Sistemas Liquido-Liquido, Um novo Algoritmo com Potencialidades de Aplicacao em Controlo. PhD thesis, Universidade do Minho, Portugal; 1995].

Automatic Classification of Anuran Sounds Using Convolutional Neural Networks

Anurans (frogs or toads) are closely related to the ecosystem and they are commonly used by biologists as early indicators of ecological stress. Automatic classification of anurans, by processing their calls, helps biologists analyze the activity of anurans on larger scale. Wireless Sensor Networks (WSNs) can be used for gathering data automatically over a large area. WSNs usually set restrictions on computing and transmission power for extending the networks lifetime. Deep Learning algorithms have gathered a lot of popularity in recent years, especially in the field of image recognition. Being an eager learner, a trained Deep Learning model does not need a lot of computing power and could be used in hardware with limited resources. This paper investigates the possibility of using Convolutional Neural Networks with Mel-Frequency Cepstral Coefficients (MFCCs) as input for the task of classifying anuran sounds.

Classifying heart sounds using SAX motifs, random forests and text mining techniques

In this paper we describe an approach to classifying heart sounds (classes Normal, Murmur and Extra-systole) that is based on the discretization of sound signals using the SAX (Symbolic Aggregate Approximation) representation. The ability of automatically classifying heart sounds or at least support human decision in this task is socially relevant to spread the reach of medical care using simple mobile devices or digital stethoscopes. In our approach, sounds are first pre-processed using signal processing techniques (decimate, low-pass filter, normalize, Shannon envelope). Then the pre-processed symbols are transformed into sequences of discrete SAX symbols. These sequences are subject to a process of motif discovery. Frequent sequences of symbols (motifs) are adopted as features. Each sound is then characterized by the frequent motifs that occur in it and their respective frequency. This is similar to the term frequency (TF) model used in text mining. In this paper we compare the TF model with the application of the TFIDF (Term frequency - Inverse Document Frequency) and the use of bi-grams (frequent size two sequences of motifs). Results show the ability of the motifs based TF approach to separate classes and the relative value of the TFIDF and the bi-grams variants. The separation of the Extra-systole class is overly difficult and much better results are obtained for separating the Murmur class. Empirical validation is conducted using real data collected in noisy environments. We have also assessed the cost-reduction potential of the proposed methods by considering a fixed cost model and using a cost sensitive meta algorithm.

Heart sounds classification using motif based segmentation

In this paper we describe an algorithm for heart sound classification (classes Normal, Murmur and Extrasystole) based on the discretization of sound signals using the SAX (Symbolic Aggregate Approximation) representation. The general strategy is to automatically discover relevant top frequent motifs and relate them with the occurrence of systolic (S1) and diastolic (S2) sounds in the audio signals. The algorithm was tuned using motifs generated from a collection of audio signals obtained from a clinical trial in a hospital. Validation was performed on a separate set of unlabeled audio signals. Results indicate ability to improve the precision of the classification of the classes Normal and Murmur.

A Parallel Algorithm for the Simulation of the Dynamic Behaviour of Liquid-Liquid Agitated Columns

Simulation of the dynamic behaviour of liquid-liquid systems is of prominent importance in many industrial fields. Algorithms for fast and reliable simulation of single stirred vessels and extraction columns have already been published by some of the present authors. In this work, we propose a methodology to develop a parallel version of a previously validated sequential algorithm, for the simulation of a liquid-liquid Kuhni column. We also discuss the algorithm implementation in a distributed memory parallel-computing environment, using MPI. Despite the difficulties encountered to preserve efficiency in the case of a heterogeneous cluster, the results demonstrate performance improvements that clearly indicate that the approach followed may be successfully extended to allow real-time plant control applications.

Studying Programming Students Motivation using Association Rules.

For Programming teachers it is of utter most importance to understand the factors that impact on students’ motivation to improve their ability to become good computer programmers. To understand a problem, to develop an algorithm for its solution, and to write the corresponding program is a challenging and arduous task, demanding time and self-confidence. In previous work we studied computer based technics to engage students in the learning activity; visualization, animation, automatic program assessment were some approaches that we combined. To support that work we studied carefully students’ motivation and complemented that study with an inquiry to a group of students of Algorithm and Programming course of the first year of an Engineering degree. In this paper we show how Association Rules can be used to mine the data gathered in the inquiry to discover relationships among factors influencing extrinsic motivation.

A Computer Platform to Increase Motivation in Programming Students - PEP.

Motivate students is one of the biggest challenges that teachers have to face, in general and in particular in programming courses. In this article two techniques, aimed at supporting the teaching of programming, are discussed: program animation, and automatic evaluation of programs. Based on the combination of these techniques and their currently available tools, we will describe two possible approaches to increase motivation and improve the success. The conclusions of a first experiment conducted in the classroom will be presented. PEP, a Web-based tool that implements one of the approaches proposed, will be introduced.

Using Smartphones to Classify Urban Sounds

The aim of this work is to develop an application for Android able to classifying urban sounds in a real life context. It also enables the collection and classification of new sounds. To train our classifier we use the UrbanSound8K data set available online. We have used a hybrid approach to obtain features, by combining SAX-based multiresolution motif discovery with Mel-Frequency Cepstral Coefficients (MFCC). We also describe different configurations of motif discovery for defining attributes and compare the use of Random Forest and SVM algorithms on this kind of data.

Animation and automatic evaluation in supporting the teaching of programming

Learning programming is a complex task that poses significant challenges. Students face different kinds of difficulties at complex levels that traditional teaching/learning methods are nor able to cope with. For this reason, several authors have researched the pedagogical effectiveness of program visualization and animation, and developed some tools. Animation can help students on the analysis and understanding of given programs, and can also guide on the development of new ones. It is very important to give students the opportunity to practice solving programming exercises by themselves. Receiving feedback is essential for knowledge acquisition. New tools arose (especially in the area of programming contests) to allow for the submission of solutions (programs developed by the students) to the problem statements presented by the teacher and to assess them, returning immediately information about the submitted answer. These tools can be incorporated into teaching activities, allowing students to test their work getting immediate feedback. Automatic evaluation systems significantly improve students performance. In this article are shown these two approaches, animation and automatic assessment, and proposed a new pedagogical practice resulting from the combination of both.