Javier Andreu-Perez

Big Data for Health

This paper provides an overview of recent developments in big data in the context of biomedical and health informatics. It outlines the key characteristics of big data and how medical and health informatics, translational bioinformatics, sensor informatics, and imaging informatics will benefit from an integrated approach of piecing together different aspects of personalized information from a diverse range of data sources, both structured and unstructured, covering genomics, proteomics, metabolomics, as well as imaging, clinical diagnosis, and long-term continuous physiological sensing of an individual. It is expected that recent advances in big data will expand our knowledge for testing new hypotheses about disease management from diagnosis to prevention to personalized treatment. The rise of big data, however, also raises challenges in terms of privacy, security, data ownership, data stewardship, and governance. This paper discusses some of the existing activities and future opportunities related to big data for health, outlining some of the key underlying issues that need to be tackled.

Deep learning for health informatics

With a massive influx of multimodality data, the role of data analytics in health informatics has grown rapidly in the last decade. This has also prompted increasing interests in the generation of analytical, data driven models based on machine learning in health informatics. Deep learning, a technique with its foundation in artificial neural networks, is emerging in recent years as a powerful tool for machine learning, promising to reshape the future of artificial intelligence. Rapid improvements in computational power, fast data storage, and parallelization have also contributed to the rapid uptake of the technology in addition to its predictive power and ability to generate automatically optimized high-level features and semantic interpretation from the input data. This article presents a comprehensive up-to-date review of research employing deep learning in health informatics, providing a critical analysis of the relative merit, and potential pitfalls of the technique as well as its future outlook. The paper mainly focuses on key applications of deep learning in the fields of translational bioinformatics, medical imaging, pervasive sensing, medical informatics, and public health.

From Wearable Sensors to Smart Implants-–Toward Pervasive and Personalized Healthcare

Objective: This paper discusses the evolution of pervasive healthcare from its inception for activity recognition using wearable sensors to the future of sensing implant deployment and data processing. Methods: We provide an overview of some of the past milestones and recent developments, categorized into different generations of pervasive sensing applications for health monitoring. This is followed by a review on recent technological advances that have allowed unobtrusive continuous sensing combined with diverse technologies to reshape the clinical workflow for both acute and chronic disease management. We discuss the opportunities of pervasive health monitoring through data linkages with other health informatics systems including the mining of health records, clinical trial databases, multiomics data integration, and social media. Conclusion: Technical advances have supported the evolution of the pervasive health paradigm toward preventative, predictive, personalized, and participatory medicine. Significance: The sensing technologies discussed in this paper and their future evolution will play a key role in realizing the goal of sustainable healthcare systems.

Multi-sensor Fusion

In the previous chapters, we have discussed issues concerning hardware, communication and network topologies for the practical deployment of Body Sensor Networks (BSNs). The pursuit of low power miniaturised distributed sensing under a patient’s natural physiological conditions has also imposed significant technical challenges on integrating information from what is often heterogeneous, incomplete and error-prone sensor data. For BSNs, the nature of errors can be attributed to a number of sources; but motion artefacts, inherent limitations and possible malfunctions of the sensors along with communication errors are the main causes of concern. In practice, it is desirable to rely on sensors with redundant or complementary data to maximise the information content and reduce both systematic errors and random artefacts. This, in essence, is the main drive for multi-sensor fusion, which is concerned with the synergistic use of multiple sources of information.

Disparity in Frontal Lobe Connectivity on a Complex Bimanual Motor Task Aids in Classification of Operator Skill Level

Objective metrics of technical performance (e.g., dexterity, time, and path length) are insufficient to fully characterize operator skill level, which may be encoded deep within neural function. Unlike reports that capture plasticity across days or weeks, this articles studies long-term plasticity in functional connectivity that occurs over years of professional task practice. Optical neuroimaging data are acquired from professional surgeons of varying experience on a complex bimanual coordination task with the aim of investigating learning-related disparity in frontal lobe functional connectivity that arises as a consequence of motor skill level. The results suggest that prefrontal and premotor seed connectivity is more critical during naïve versus expert performance. Given learning-related differences in connectivity, a least-squares support vector machine with a radial basis function kernel is employed to evaluate skill level using connectivity data. The results demonstrate discrimination of operator skill level with accuracy ≥0.82 and Multiclass Matthews Correlation Coefficient ≥0.70. Furthermore, these indices are improved when local (i.e., within-region) rather than inter-regional (i.e., between-region) frontal connectivity is considered (p = 0.002). The results suggest that it is possible to classify operator skill level with good accuracy from functional connectivity data, upon which objective assessment and neurofeedback may be used to improve operator performance during technical skill training.

A Self-Adaptive Online Brain–Machine Interface of a Humanoid Robot Through a General Type-2 Fuzzy Inference System

This paper presents a self-adaptive autonomous online learning through a general type-2 fuzzy system (GT2 FS) for the motor imagery (MI) decoding of a brain-machine interface (BMI) and navigation of a bipedal humanoid robot in a real experiment, using electroencephalography (EEG) brain recordings only. GT2 FSs are applied to BMI for the first time in this study. We also account for several constraints commonly associated with BMI in real practice: 1) the maximum number of EEG channels is limited and fixed; 2) no possibility of performing repeated user training sessions; and 3) desirable use of unsupervised and low-complexity feature extraction methods. The novel online learning method presented in this paper consists of a self-adaptive GT2 FS that can autonomously self-adapt both its parameters and structure via creation, fusion, and scaling of the fuzzy system rules in an online BMI experiment with a real robot. The structure identification is based on an online GT2 Gath–Geva algorithm where every MI decoding class can be represented by multiple fuzzy rules (models), which are learnt in a continous (trial-by-trial) non-iterative basis. The effectiveness of the proposed method is demonstrated in a detailed BMI experiment, in which 15 untrained users were able to accurately interface with a humanoid robot, in a single session, using signals from six EEG electrodes only.

EALab (Eye Activity Lab): a MATLAB Toolbox for Variable Extraction, Multivariate Analysis and Classification of Eye-Movement Data

Recent advances in the reliability of the eye-tracking methodology as well as the increasing availability of affordable non-intrusive technology have opened the door to new research opportunities in a variety of areas and applications. This has raised increasing interest within disciplines such as medicine, business and education for analysing human perceptual and psychological processes based on eye-tracking data. However, most of the currently available software requires programming skills and focuses on the analysis of a limited set of eye-movement measures (e.g., saccades and fixations), thus excluding other measures of interest to the classification of a determined state or condition. This paper describes ‘EALab’, a MATLAB toolbox aimed at easing the extraction, multivariate analysis and classification stages of eye-activity data collected from commercial and independent eye trackers. The processing implemented in this toolbox enables to evaluate variables extracted from a wide range of measures including saccades, fixations, blinks, pupil diameter and glissades. Using EALab does not require any programming and the analysis can be performed through a user-friendly graphical user interface (GUI) consisting of three processing modules: 1) eye-activity measure extraction interface, 2) variable selection and analysis interface, and 3) classification interface.

A virtual reality and brain computer interface system for upper limb rehabilitation of post stroke patients

This work presents a brain computer interface (BCI) framework for upper limb rehabilitation of post stroke patients, combining BCI and virtual reality (VR) technology; a VR feedback is shown to the participants to achieve a greater activation of certain brain regions involved with the performing of upper limb motor task. This system uses an adaptive neuro-fuzzy inference system (ANFIS) classifier to discriminate between a motor task and rest condition, the first one classifies between extension and rest conditions; and the second one classifies between flexion and rest conditions. In the training stage, eight healthy subjects participated in the sessions, the best accuracies are 99.3% and 88.9%, as a result of cross-validation. Meanwhile, the best accuracy in online test is 89%. The methodology here presented can be straightforwardly employed as a rehabilitation system for brain repair in individuals with neurological diseases or brain injury.

Improved estimation of effective brain connectivity in functional neuroimaging through higher order fuzzy cognitive maps

In this paper, a novel technique for the computation of effective brain connectivity in functional Near-Infrared Spectroscopy (fNIRS) data is presented. The estimation of effective brain connectivity using the proposed approach of higher order Fuzzy Cognitive Maps (FCMs), used in conjunction with Genetic Algorithm (GA), is shown to be more accurate. Owing to lack of dependency on human knowledge, the FCM-GA model becomes more robust to subjective beliefs of experts from various domains when establishing connectivity matrix. Furthermore, higher order FCMs are capable of assessing causal relations in historical data with variable time lag, g, therefore generating more accurate predictions for complex causal data such as fNIRS where the causality may not necessarily follow a first order dynamics. The computation model of higher order FCM-GA is shown to perform better than Granger Causality (GC) for estimating effective brain connectivity in synthetic fNIRS data at 95% significance level. The proposed approach is also tested on real fNIRS data, and shown to estimate the causal structure amongst region of interests (ROIs) with improved accuracy.

A P300-based brain computer interface for smart home interaction through an ANFIS ensemble

Adaptive neuro fuzzy Inference systems (ANFIS) has been applied in brain computer interfaces (BcI) in different ways such as mapping of P300 or fusing information from EEG channels and it has reached high classification accuracy. This work proposes a combination of ANFIS classifiers by voting for a single-trial detection of a P300 wave in a BCI, using four channels; five healthy subjects and three post-stroke patients have participated in this study, each participant performs 4 BCI sessions, crossvalidation is applied to evaluate the classifier performance. The results of average accuracy were greater than 75% for all subjects, similar results were gotten for healthy subjects and post-stroke patients, but the better classifiers for each subject have achieved accuracies greater than 80%.