Image Processing Including Medical Liver Imaging

Medical diagnosis has been gaining importance in everyday life. The diseases and their symptoms are highly varying and there is always a need for a continuous update of knowledge needed for the doctors. The diseases fall into different categories and a small variation of symptoms may leave to different categories of diseases. This is further supplemented by the medical analysts for a continuous treatment process. The treatment generally starts with a diagnosis and further goes through a set of procedures including X-ray, CT-scans, ultrasound imaging for qualitative analysis and diagnosis by doctors. A small level of error in disease identification introduces overhead in diagnosis and difficult in treatment. In such cases, an automated system that could retrieve medical images based on user's interest. This chapter deals with various techniques, methodologies that correspond to the classification problem in data analysis process and its methodological impacts to big data.

Applying Bayesian Networks in the Early Diagnosis of Bulimia and Anorexia Nervosa in Adolescents

The diseases and health problems are concerns of managers of the Unified Health System has costs in more sophisticated care sector are high. The World Health Organization focused on prevention of chronic diseases to prevent millions of premature deaths in the coming years, bringing substantial gains in economic growth by improving the quality of life. Few countries appear to be aimed at prevention, if not note the available knowledge and control of chronic diseases and may represent an unnecessary risk to future generations. Early diagnosis of these diseases is the first step to successful treatment in any age group. The objective is to build a model, from the establishment of a Bayesian network, for the early diagnosis of nursing to identify eating disorders bulimia and anorexia nervosa in adolescents, from the characteristics of the DSM-IV and Nursing Diagnoses The need for greater investment in technology in public health actions aims to increase the knowledge of health professionals, especially nurses, contributing to prevention, decision making and early treatment of problems.

Selection of Pathway Markers for Cancer Using Collaborative Binary Multi-Swarm Optimization

Pathway information for cancer detection helps to find co-regulated gene groups whose collective expression is strongly associated with cancer development. In this paper, a collaborative multi-swarm binary particle swarm optimization (MS-BPSO) based gene selection technique is proposed that outperforms to identify the pathway marker genes. We have compared our proposed method with various statistical and pathway based gene selection techniques for different popular cancer datasets as well as a detailed comparative study is illustrated using different meta-heuristic algorithms like binary coded particle swarm optimization (BPSO), binary coded differential evolution (BDE), binary coded artificial bee colony (BABC) and genetic algorithm (GA). Experimental results show that the proposed MS-BPSO based method performs significantly better and the improved multi swarm concept generates a good subset of pathway markers which provides more effective insight to the gene-disease association with high accuracy and reliability.

An Optimized Semi-Supervised Learning Approach for High Dimensional Datasets

The semi-supervised learning is one of the most interesting fields for research developments in the machine learning domain beyond the scope of supervised learning from data. Medical diagnostic process works mostly in supervised mode, but in reality, we are in the presence of a large amount of unlabeled samples and a small set of labeled examples characterized by thousands of features. This problem is known under the term “the curse of dimensionality”. In this study, we propose, as solution, a new approach in semi-supervised learning that we would call Optim Co-forest. The Optim Co-forest algorithm combines the re-sampling data approach (Bagging Breiman, 1996) with two selection strategies. The first one involves selecting random subset of parameters to construct the ensemble of classifiers following the principle of Co-forest (Li & Zhou, 2007). The second strategy is an extension of the importance measure of Random Forest (RF; Breiman, 2001). Experiments on high dimensional datasets confirm the power of the adopted selection strategies in the scalability of our method.

Protein Structure Prediction

The great disagreement between the number of known protein sequences and the number of experimentally determined protein structures indicate an enormous necessity of rapid and accurate protein structure prediction methods. Computational techniques such as comparative modeling, threading and ab initio modelling allow swift protein structure prediction with sufficient accuracy. The three phases of computational protein structure prediction comprise: the pre-modelling analysis phase, model construction and post-modelling refinement. Protein modelling is primarily comparative or ab initio. Comparative or template-based methods such as homology and threading-based modelling require structural templates for constructing the structure of a target sequence. The ab initio is a template-free modelling approach which proceeds by satisfying various physics-based and knowledge-based parameters. The chapter will elaborate on the three phases of modelling, the programs available for performing each, issues, possible solutions and future research areas.

Transcriptomics to Metabolomics

A lot of omics data is generated in a recent decade which flooded the internet with transcriptomic, genomics, proteomics and metabolomics data. A number of software, tools, and web-servers have developed to analyze the big data omics. This review integrates the various methods that have been employed over the years to interpret the gene regulatory and metabolic networks. It illustrates random networks, scale-free networks, small world network, bipartite networks and other topological analysis which fits in biological networks. Transcriptome to metabolome network is of interest because of key enzymes identification and regulatory hub genes prediction. It also provides an insight into the understanding of omics technologies, generation of data and impact of in-silico analysis on the scientific community.

Bioinformatics

This chapter describes the overview of bioinformatics; bioinformatics, data mining, and data visualization; bioinformatics and secretome analysis; bioinformatics, mass spectrometry, and chemical cross-linking reagents; bioinformatics and Software Product Line (SPL); bioinformatics and protein kinase; bioinformatics and MicroRNAs (miRNAs); and clinical bioinformatics and cancer. Bioinformatics is the application of computer technology to the management and analysis of biological data. Bioinformatics is an interdisciplinary research area that is the interface between biology and computer science. The primary goal of bioinformatics is to reveal the wealth of biological information hidden in the large amounts of data and obtain a clearer insight into the fundamental biology of organisms. Bioinformatics entails the creation and advancement of databases, algorithms, computational and statistical techniques, and theory to solve the formal and practical problems arising from the management and analysis of biological data.

Effective and Efficient Business Intelligence Dashboard Design

This research focuses on the design process of an effective and efficient dashboard which displays management information for an Electronic Health Record (EHR) in Dutch long-term and chronic healthcare. It presents the actual design and realization of a management dashboard for the YBoard 2.0 system, which is a popular solution on the Dutch market. The design decisions in this investigation were based on human perception and computer interaction theory, in particular Gestalt theory. The empirical interviews with medical professionals supplemented valuable additional insights into what the users wanted to see most of all in a dashboard in their daily practices. This study successfully shows how effective and efficient dashboard design can benefit from theoretical insights related to human perception and computer interaction such as Gestalt theory, in combination with integrated end user requirements from daily practices.

Protein Docking and Drug Design

Protein docking is integral to structure-based drug design and molecular biology. The recent surge of big data in biology, the demand for personalised medicines, evolving pathogens and increasing lifestyle-associated risks, asks for smart, robust, low-cost and high-throughput drug design. Computer-aided drug design techniques allow rapid screening of ultra-large chemical libraries within minutes. This is immensely necessary to the drug discovery pipeline, which is presently burdened with high attrition rates, failures, huge capital and time investment. With increasing drug resistance and difficult druggable targets, there is a growing need for novel drug scaffolds which is partly satisfied by fragment based drug design and de novo methods. The chapter discusses various aspects of protein docking and emphasises on its application in drug design.

Informatics and Data Analytics to Support Exposome-Based Discovery

This chapter aims at outlining the current state of science in the field of computational exposure biology and in particular at demonstrating how the bioinformatics techniques and algorithms can be used to support the association between environmental exposures and human health and the deciphering of the molecular and metabolic pathways of induced toxicity related to environmental chemical stressors. Examples of the integrated bioinformatics analyses outlined herein are given concerning exposure to airborne chemical mixtures, to organic compounds frequently found in consumer goods, and to mixtures of organic chemicals and metals through multiple exposure pathways. Advanced bioinformatics are coupled with big data analytics to perform studies of exposome-wide associations with putative adverse health outcomes. In conclusion, the chapter gives the reader an outline of the available computational tools and paves the way towards the development of future comprehensive applications that are expected to support efficiently exposome research in the 21st century.