Ebook: Innovation in Medicine and Healthcare 2014

The second KES International Conference on Innovation in Medicine and Healthcare (InMed-14) was held over 9–11 July 2014 in San Sebastian, Spain, organised by KES International in partnership with the University of the Basque Country, UPV/EHU and the Institute of Knowledge Transfer.

InMed-14 gathered researchers and engineers, managers, students and practitioners from a broad medically-related arena. This multi-disciplinary group met to discuss the ways that technological and methodological innovation, knowledge exchange and enterprise can be applied to issues relating to medicine, surgery, healthcare and the issues of an ageing population. A central theme of the conference was Smart Medical and Healthcare Systems which covered the ways in which modern intelligent systems contribute to the solution of problems faced by healthcare and medical practitioners today, addressing the application of these systems.

The conference featured five excellent keynote talks from internationally renowned experts, namely:

• Edward J. Ciaccio, Columbia University, NY (talk entitled ‘Model of Reentrant Ventricular Tachycardia based on Wavefront Curvature’);

• Jesus Cortes, Biocruces, Ikerbasque, Spain (talk entitled ‘Computational Neuroimaging for Health and Disease’);

• Juan Manuel Gorriz, University of Granada, Spain (talk entitled ‘DiagnoSIS: Diagnosis by means of Statistical and Intelligent Systems’);

• Ricardo Sanchez Peña, ITBA & CONICET, Argentina (talk entitled ‘Automatic Control of Diabetes type 1’); and

• Sebastiano Stramaglia, Ikerbasque Visiting Professor, Spain (talk entitled ‘Causality measures for Brain Computation’).

In addition to the General Track, chaired by Prof Manuel Grana, University of the Basque Country UPV/EHU and the Workshop on Smart Medical and Healthcare, chaired by Dr Carlos Toro, Vicomtech-IK4, Systems, there were special sessions on Medical Decision-Support Systems chaired by Dr Elena Hernández-Pereira, University of La Coruña, Spain; Computer-aided Image Analyis in Ophthalmology chaired by Prof Manuel Penedo, University of La Coruña, Spain; and Advances in Data & Knowledge Management for Healthcare chaired by Dr Massimo Esposito, National Research Council of Italy (ICAR-CNR).

These proceedings consist of 43 papers that were presented at the conference, each of which was comprehensively reviewed by at least two members of the International Programme Committee.

We hope this will form a useful and interesting reference for further research on this topic.

The InMed-14 Conference Chairs:

Manuel Graña, Robert J. Howlett, Lakhmi C. Jain and Carlos Toro

In this work we present a system that uses the accelerometer embedded in a mobile phone to perform activity recognition, with the purpose of continuously and pervasively monitoring the users' level of physical activity in their everyday life. Several classification algorithms are analysed and their performance measured, based for 6 different activities, namely walking, running, climbing stairs, descending stairs, sitting and standing. Feature selection has also been explored in order to minimize computational load, which is one of the main concerns given the restrictions of smartphones in terms of processor capabilities and specially battery life.

Alzheimer's Diasese (AD) diagnosis can be carried out by analysing functional or structural changes in the brain. Functional changes associated to neurological disorders can be figured out by positron emission tomography (PET) as it allows to study the activation of certain areas of the brain during specific task development. On the other hand, neurological disorders can also be discovered by analysing structural changes in the brain which are usually assessed by Magnetic Resonance Imaging (MRI). In fact, computer-aided diagnosis tools (CAD) that have been recently devised for the diagnosis of neurological disorders use functional or structural data. However, functional and structural data can be fused out in order to improve the accuracy and to diminish the false positive rate in CAD tools. In this paper we present a method for the diagnosis of AD which fuses multimodal image (PET and MRI) data by combining Sparse Representation Classifiers (SRC). The method presented in this work shows accuracy values up to 95% and clearly outperforms the classification outcomes obtained using single-modality images.

This paper presents the analysis of the statistical significance in the selection of the ROI for the discriminant analysis of brain images to identify Parkinson patients or subjects without any pathology. The particular features and brain functional patterns of the Parkinson's disease cause that there are regions that conveniently reveal the presence of the pathology, in this case mainly the striatum region. The selection of the brain mask makes incidence in two main aspects: the selection of the region of interest (striatum and surrounding area) for the analysis, but also the selection of the region without significance, which is the reference area for the intensity normalization, previous to the analysis. This work studies the statistical significance in the selection of ROIs in 3D brain images for Parkinson, depending on the objective to be achieved in the posterior analysis process.

Medical images are being studied to analyse the brain in neurological disorders. Measurements extracted from Diffusion tensor image (DTI) such as Fractional Anisotropy (FA) describe the brain changes caused by diseases. However, there is no single best method for the quantitative brain analysis. This paper presents a review of the existing methods and software tools for brain analysis through DTI measurements. It also states some challenges that current software tools still have to meet in order to improve automation and usability and become smarter software tools.

Human body motion is usually variable in terms of intensity and, therefore, any Inertial Measurement Unit attached to a subject will measure both low and high angular rate and accelerations. This can be a problem for the accuracy of orientation estimation algorithms based on adaptive filters such as the Kalman filter, since both the variances of the process noise and the measurement noise are set at the beginning of the algorithm and remain constant during its execution. Setting fixed noise parameters burdens the adaptation capability of the filter if the intensity of the motion changes rapidly. In this work we present a conjoint novel algorithm which uses a motion intensity detector to dynamically vary the noise statistical parameters of different approaches of the Kalman filter. Results show that the precision of the estimated orientation in terms of the RMSE can be improved up to 29% with respect to the standard fixed-parameters approaches.

Optical Coherence Tomography (OCT) is a promising imaging technique used by ophthalmologists to diagnose diseases. Since retinal morphology can be identified on these images, several image processing-based methods are emerging with the purpose of extracting their information. The first step to tackle any automatic method to extract pathological features from these images is delimiting retinal layers automatically. This is the aim of this paper, which presents an active contour-based method to segment layer boundaries in the retina. Results obtained by this method present high accuracy and robustness, even when some of these layers are low defined or vessel shades are present.

The tortuosity of a vessel, that is, how many times a vessel curves, and how these turns are, is an important value for the diagnosis of certain diseases. Clinicians analyze fundus images manually in order to estimate it, but there is many drawbacks as it is a tedious, time-consuming and subjective work. Thus, automatic image processing methods become a necessity, as they make possible the efficient computation of objective parameters. In this paper we will discuss Sirius (System for the Integration of Retinal Images Understanding Service), a web-based application that enables the storage and treatment of various types of diagnostic tests and, more specifically, its tortuosity calculation module.

In this work, we perform a comparison between the spatial normalization of [123I]FP-CIT SPECT brain images when a FP-CIT SPECT and a MRI template are used. A 12-parameters affine registration model is calculated by the optimization of a sum of squares cost function. When the images are registered to a FP-CIT template, the intersubject variation is found to be lower than when the MRI template is used, specially in the striatum, which is the most relevant part of the brain in FP-CIT SPECT brain images.

In this paper we investigate a new approach for extracting features from a texture using Dijkstra's algorithm. The method maps images into graphs and gray level differences into transition costs. Texture is measured over the whole image comparing the costs found by Dijkstra's algorithm with the geometric distance of the pixels. In addition, we compare and combine our new strategy with a previous method for describing textures based on Dijkstra's algorithm. For each set of features, a support vector machine (SVM) is trained. The set of classifiers is then combined by weighted sum rule. Combining the proposed set of features with the well-known local binary patterns and local ternary patterns boosts performance. To assess the performance of our approach, we test it using six medical datasets representing different image classification problems. Tests demonstrate that our approach outperforms the performance of standard methods presented in the literature. All source code for the approaches tested in this paper will be available at: http://www.dei.unipd.it/node/2357.

In this paper we propose an ensemble of texture descriptors for analyzing virus textures in transmission electron microscopy images. Specifically, we present several novel multi-quinary (MQ) codings of local binary pattern (LBP) variants: the MQ version of the dense LBP, the MQ version of the rotation invariant co-occurrence among adjacent LBPs, and the MQ version of the LBP histogram Fourier. To reduce computation time as well as to improve performance, a feature selection approach is utilized to select the thresholds used in the MQ approaches. In addition, we propose new variants of descriptors where two histograms, instead of the standard one histogram, are produced for each descriptor. The two histograms (one for edge pixels and the other for non-edge pixels) are calculated for training two different SVMs, whose results are then combined by sum rule. We show that a bag of features approach works well with this problem. Our experiments, using a publicly available dataset of 1500 images with 15 classes and same protocol as in previous works, demonstrate the superiority of our new proposed ensemble of texture descriptors. The MATLAB code of our approach is available at https://www.dei.unipd.it/node/2357.

One of the main challenges of classifying clinical data is determining how to handle missing features. Most research favours imputing of missing values or neglecting records that include missing data, both of which can degrade accuracy when missing values exceed a certain level. In this research we propose a methodology to handle data sets with a large percentage of missing values and with high variability in which particular data are missing. Feature selection is effected by picking variables sequentially in order of maximum correlation with the dependent variable and minimum correlation with variables already selected. Classification models are generated individually for each test case based on its particular feature set and the matching data values available in the training population.

The method was applied to real patients' anonymous mental-health data where the task was to predict the suicide risk judgement clinicians would give for each patient's data, with eleven possible outcome classes: zero to ten, representing no risk to maximum risk. The results compare favourably with alternative methods and have the advantage of ensuring explanations of risk are based only on the data given, not imputed data. This is important for clinical decision support systems using human expertise for modelling and explaining predictions.

Distance education has grown in importance with the advent of the internet. An adequate evaluation of students in this mode is still difficult. Distance tests or occasional on-site exams do not meet the needs of evaluation of the learning process for distance education. Bayesian networks are adequate for simulating several aspects of clinical reasoning. The possibility of integrating them in distance education student evaluation has not yet been explored much. The present work describes a Simulator based on probabilistic networks built to represent knowledge of clinical practice guidelines in Family and Community Medicine. The Bayesian Network, the basis of the simulator, was modeled to playable by the student, to give immediate feedback according to pedagogical strategies adapted to the student according to past performance, and to give a broad evaluation of performance at the end of the game. Simulators structured by Bayesian Networks may become alternatives in the evaluation of students of Medical Distance Education.

This paper presents REMPARK system, a novel approach to deal with Parkinson's Disease (PD). REMPARK system comprises two closed loops of actuation onto PD. The first loop consists in a wearable system that, based on a belt-worn movement sensor, detects movement alterations that activate an auditory cueing system controlled by a smartphone in order to improve patient's gait. The belt-worn sensor analyzes patient's movement through real-time learning algorithms that were developed on the basis of a database previously collected from 93 PD patients. The second loop consists in disease management based on the data collected during long periods and that enables neurologists to tailor medication of their PD patients and follow the disease evolution. REMPARK system is going to be tested in 40 PD patients in Spain, Ireland, Italy and Israel. This paper describes the approach followed to obtain this system, its components, functionalities and trials in which the system will be validated.

With recent developments in wireless networks field and in sensing technology, new and innovative medical applications based on the wireless sensor networks (WSN) technology are being developed in the research lab or commercial sectors. The key problem of WSN based application in the medical and health care domain is to get the message with the bounded delays and/or real-time data such as the signal of falls, the alarm of heart attacks traversed to the cluster at where the emergence unit located while delivering the steady flows of data with a certain level of Quality of Service (QoS). In this paper, we address this problem by proposing to use a periodic scheduling–Time Division Cluster Schedule (TDCS), to avoid the flow collision while meeting the end-to-end delay deadlines. The proposed scheduling mechanism has been studied on simulated Wireless sensor networks with a cluster tree topology on which the OPEN-ZB simulation model is adopted. The preliminary results shown that it is feasible to guarantee the real time medical data delivered on time over the low data rate wireless sensor networks when deployed at medical and/or health care domain, under the governance of the TDCS mechanism.

The mechanical behavior of stents is one of the important factors involved in ensuring their function in maintaining an open blood vessel. This study aims to study the development of different kinds of fiber-based stents, using braiding technology. Moreover, the impact of braiding angle and mandrel's diameter in the mechanical behavior of the stent is also analyzed. Furthermore, stent's mechanical properties like radial and longitudinal compression, bending and porosity will be measured and discuss. The results show that fibrous stents present suitable mechanical properties, when compared to those of commercial ones, and may reduce the disadvantages of commercial metallic stents.

Recently the indicators characterizing health state development of population in Armenia have undergone various negative and structural changes.

To soften the negative influence of indicators in 2011 “Child's Health State Certificate” program was implemented. Consequently sharp rise of the number of served cases was detected.

To abolish the existing defects funding of completed cases and appropriate payment provision should be done in outpatient service.

Mandatory conditions for completed cases are diagnosis, appropriate laboratory, instrumental investigations, treatment prescription, approprate supervision, treatment outcome: cure, amelioration, medical treatment referral absence during the sequential 5 days with the same diagnosis.

Usage of the mentioned mechanism will make direct and visible connection between the amounts of services provided by doctors and received salary, which will boost doctors' work efficiency and make conditions to reduce groundless referrals to medical institutions.

Using game technologies and digital media for improving physical and mental health and for the therapeutic benefit and well-being of a wide range of people is an area of study that is rapidly expanding. Much research in this emerging field is centered at the intersection of serious games, alternative realities, and play therapy. In this paper the authors describe their transdisciplinary work at this intersection: i) an integrative system of psychotherapy technologies called MyPsySpace currently being prototyped in Second Life with the aim of offering new and virtual translations of traditional expressive therapies (virtual sandplay, virtual drama therapy, digital expressive therapy, and virtual safe spaces) and ii) a mature body of research entitled SoundScapes that is exploring the use of interactive video games and abstract creative expression (making music, digital painting, and robotic device control) as a supplement to traditional physical rehabilitation intervention. Aside from introducing our work to a broader audience, our goal is to encourage peers to investigate ideas that reach across disciplines–to both risk and reap the benefits of combining technologies, theories, and methods stemming from multiple disciplines.

Generally, current clinical imaging methods do not provide highly detailed information about location of axonal injury, severity of injury or expected recovery of patients with traumatic brain injury (TBI). High-Definition Fiber Tractography (HDFT) is a novel imaging modality that allows visualizing and quantifying, directly, the degree of axons damage, predicting functional deficits due to traumatic axonal injury and loss of cortical projections. This imaging modality is based on diffusion technology. Being a novel modality, validation and quality control are essential. Thus this study aims at the development of a brain phantom to mimic the human brain in order to fill some gaps that currently exist in this area. This paper is focused on this novel imaging approach, the role of brain phantoms on its validation and the quality control, as well as, on the materials used in their construction. Furthermore, some important characteristics of fibrous materials for brain phantom are also discussed.

Dry eye syndrome is a common disorder of the tear film which affects a remarkable percentage of the population. The Break-Up Time (BUT) is a clinical test used for the diagnosis of this disease, which computes the time the first tear film break-up appears. This work describes a fully automatic methodology to compute the BUT measurement and evaluate the break-up dynamics until the final blink. This analysis provides useful additional information for the assessment of tear film stability.

Accurate resuscitation of the critically-ill patient using intravenous fluids and blood products is a challenging, time sensitive task. Ultrasound of the inferior vena cava (IVC) is a non-invasive technique currently used to guide fluid administration, though multiple factors such as variable image quality, time, and operator skill challenge mainstream acceptance. This study represents a first attempt to develop and validate an algorithm capable of automatically tracking and measuring the IVC compared to human operators across a diverse range of image quality. Minimal tracking failures and high levels of agreement between manual and algorithm measurements were demonstrated on good quality videos. Addressing problems such as gaps in the vessel wall and intra-lumen speckle should result in improved performance in average and poor quality videos. Semi-automated measurement of the IVC for the purposes of non-invasive estimation of circulating blood volume poses challenges however is feasible.

Failure to detect patients at risk of attempting suicide can result in tragic consequences. Identifying risks earlier and more accurately helps prevent serious incidents occurring and is the objective of the GRiST clinical decision support system (CDSS). One of the problems it faces is high variability in the type and quantity of data submitted for patients, who are assessed in multiple contexts along the care pathway. Although GRiST identifies up to 138 patient cues to collect, only about half of them are relevant for any one patient and their roles may not be for risk evaluation but more for risk management.

This paper explores the data collection behaviour of clinicians using GRiST to see whether it can elucidate which variables are important for risk evaluations and when. The GRiST CDSS is based on a cognitive model of human expertise manifested by a sophisticated hierarchical knowledge structure or tree. This structure is used by the GRiST interface to provide top-down controlled access to the patient data. Our research explores relationships between the answers given to these higher-level “branch” questions to see whether they can help direct assessors to the most important data, depending on the patient profile and assessment context. The outcome is a model for dynamic data collection driven by the knowledge hierarchy. It has potential for improving other clinical decision support systems operating in domains with high dimensional data that are only partially collected and in a variety of combinations.

Scientific literature has been quickly expanding as the availability of articles in electronic form has increased rapidly. For the scientific researcher and the practitioner alike, keeping track with the advancement of the research is an ongoing challenge, and for the most part, the mass of experience recorded in the scientific literature is largely untapped. In particular, novice scientists, non researchers, and students would benefit from a system proposing recommendations for the problems they are interested in resolving. This article presents the first stages of the Digital Knowledge Finder design, a case-based reasoning system to manage experience from the scientific literature. One of the main functionality of the system is to enable both to represent the experience in a declarative and searchable form, and to reason from it through reuse – the latter being a consequence of the former. This article focuses on research findings mining and results from an aging literature dataset.

This paper presents a comparative study over the respiratory pattern classification task involving three missing data imputation techniques, and four different machine learning algorithms. The main goal was to find a classifier that achieves the best accuracy results using a scalable imputation method in comparison to the method used in a previous work of the authors. The results obtained show that the Self-organization maps imputation method allows any classifier to achieve improvements over the rest of the imputation methods, and that the Feedforward neural network classifier offers the best performance regardless the imputation method used.

Recent advances in the process of diagnosis of neurodegenerative diseases, such as Alzheimer's Disease, rely on the use of molecular imaging that allow the interpretation of different metabolic biomarkers in the brain. However these procedures are considered of invasive nature, as they involve the injection of radioactive markers. On the other hand, Magnetic Resonance Imaging (MRI) is perhaps the most widely used and less invasive medical imaging technique, although its ability to detect Alzheimer's Disease has revealed limited. In this paper, a new method that simplifies the process of analysing 3D MRI brain images using a two dimensional projection is proposed. Our system outperforms other methods that use MRI, achieving up to a 86% of accuracy and significantly reducing the computational load. Additionally, it allows the visual analysis and interpretation of the images, which can be of great help in the diagnosis of this and other types of dementia.