International Rural and Elderly Health Informatics Conference (IREHI) Edition 2019  Open Sessions

Prof. Samuel Fosso-Waba

Toulouse Business School



Dr. Haroldo Campos Velho

INPE, S. J. dos Campos, SP, National Institute for Space Research
Title: Precision agriculture and self-configuring neural networks

Agriculture is one of the most important sectors in the economy and essential for the population. Increasing agriculture production with low-price is a relevant task. Technological advances can help to improve the efficiency of agricultural production. Planting and harvest planning is linked to forecasting rainy and/or drought episodes. In addition, pest control in crops is also an important issue to be considered. In this talk, we will use self-configuring neural networks to perform climate seasonal precipitation prediction and autonomous navigation of unmanned aerial vehicles (UAV) without the use of global navigation satellite systems (GNSS). The use of GNSS signal may not be applicable for countries in the equatorial zone  – in the talk, it will be explained why the signal of such systems may fail.

Dr. Abdeldjalil Khelassi
Abou Bakr Belkaid

University of Tlemcen

Title: Application of Explanation-Aware Computing in Cardiac Arrhythmias Diagnosis

The cardiovascular diseases represent a big part of mortality causes. Cardiac arrhythmia is well concerned by researchers via important research towards and contributions. Several contributions attend the success with a high level of accuracy in cardiac arrhythmias recognize, but the uncertainty and risks still the disadvantage of these contributions. Explanation-aware computing, as an important domain of artificial intelligence, is promoting several aspects for resolving the weaknesses of uncertainty. In this talk, we will present our application in cardiac arrhythmias diagnosis and the explainer that we develop against uncertainty.        
Keywords: explanation-aware computing, Cardiac arrhythmia diagnosis, uncertainty
Prof. Nataliya Shakhovska

Lviv Polytechnic National University, Lviv,
Title: Association rules in Medical Domain

Association rules are a data mining technique used to discover frequent patterns in a data set. In this work, association rules are used in the medical domain, where data sets are generally high dimensional. The chief disadvantage of mining association rules in a high dimensional data set is the huge number of patterns that are discovered, most of which are irrelevant or redundant. This disadvantage is grown when Big data is used. The multidimensional view of the data is well used for data visualization and analysis tasks, but due to the hypercube dissipation, the amount of data, in this case, is greater than the relational representation that is not acceptable to the Big Data. Object representation allows you to store an object in the form of attributes, their characteristics, and relationships between characteristics. For some modification, it can be used for Big Data.
In medical and biological research, as well as in practical medicine, the range of tasks to be solved is so wide that it is possible to use any of the methodologies of Data Mining. An example can be the construction of a diagnostic system or the study of the effectiveness of a surgical intervention.
One of the most advanced areas of medicine is bioinformatics. The object of bioinformatics research is huge amounts of information about DNA sequences and the primary structure of proteins that arose as a result of studying the structure of genomes of microorganisms, mammals, and humans. Abstracted from the specific content of this information, it can be regarded as a set of genetic texts, consisting of extended character sequences. Detection of structural laws in such sequences is a number of tasks, effectively solved by means of Data Mining, for example, by means of sequencing and associative analysis.
The purpose of the study is to identify the most important rules for constructing associative rules. We should analyze not only single parameters and their values but also combining these parameters in groups. Determination of the patterns of constructing associative rules and the division of physical indicators at different levels of the hierarchy.
Prof. Solomiia Fedushko

Lviv Polytechnic National University, Lviv,
Title: Systems for monitoring physical and psychological conditions of the person in real-time

Patient health research has always remained an important issue in medicine because, in the absence of certain data, the lack of digitization of research and thorough study information has become a difficult process or impossible at all. The Real-time monitoring system of the physical, psychological state of a person is a necessary component, which will allow obtaining additional insights about the patient’s health, persons’ way of life and to prevent certain diseases; carry out an experimental analysis of data based on information. Intellectual systems, applications that contribute to a detailed time series analysis, development of metrics for assessing human health, integration and systematization of data play a major role in this process. Information gathering and consolidation should be done using a variety of sensors, internet of things devices and more. This, in turn, will greatly facilitate the detection of human health anomalies, additional research, A / B testing. Developing appropriate systems and frameworks saves time and cost while conducting various patient studies and will consolidate information and knowledge into a single, real-time monitoring service.