Community Research and Development Information Service - CORDIS

H2020

myAirCoach Report Summary

Project ID: 643607
Funded under: H2020-EU.3.1.

Periodic Reporting for period 1 - myAirCoach (Analysis, modelling and sensing of both physiological and environmental factors for the customized and predictive self-management of Asthma)

Reporting period: 2015-01-01 to 2015-12-31

Summary of the context and overall objectives of the project

What is MyAirCoach
MyAirCoach project seeks to create a patient centered mHeatlh tool to support self-management approaches for asthma. The project will enable healthcare professionals to supervise the patients’ condition in efficiently without disturbing patients’ privacy. MyAirCoach final system will stimulate and increase the asthma self-management awareness and will serve as an exchange platform for patients. MyAirCoach proposes a novel mHealth tool based on a wireless body sensor network that will be the core element of a new approach to monitor and support asthma patients.
The system will communicate in two senses: 1) to the healthcare professional by observing patients' adherence to medical treatment through physiological and environmental variables and 2) to the patient, as it will provide them with personalized prediction to manage and reduce the risk of asthma exacerbations. The system leverages the ongoing integration and miniaturization of sensors to build an integrated holistic mHealth asthma self-management framework that is expected to become an integral part of the existing clinical procedures and asthma treatment protocols.
Why
- Asthma remains uncontrolled, despite the wide availability of asthma therapies and guidelines and the latest achievements on respiratory diseases monitoring and self-management;
- Difficult long-term asthma management, as it frequently falls short on the goals set by healthcare guidelines and medical experts;
- Incorrect self-management, asthma patients might use information online to understand and treat the disease, without involving a healthcare professional;
- Asthma needs individualized attention, the optimal asthma treatment depends on managing dynamic parameters like the patient’s physiological state, behavioral factors, environmental parameters and treatment compliance;
- Asthma poses a great burden on patients, those who do not manage to achieve the targets experience a significant impact on their quality of life.
How
MyAirCoach is composed of an interdisciplinary research team that will apply the approach to patients in two measurement campaigns and three pilot sites in Europe.
More specifically, the project will focus on looking to innovate:
· computational modelling of the pulmonary system and patient-specific models based on dynamical physiological, behavioral and environmental variables,
· wireless sensors to provide up-to-date physiological, environmental and behavioral measurements,
· prediction tools to understand asthma triggers in outdoor and indoor environments,
· self-management and efficient personalized guidance and supporting mechanisms to control asthma.
For Whom
The project is expected to demonstrate significant impact in various domains ranging from the patients themselves, healthcare professionals, associations and the pharmaceutical companies. At a glance:
- Patients will receive optimal strategies to treat their asthma based on their own models, the estimated disease progression and their evolution, all by their doctor. Through MyAirCoach community, they will also be able to
communicate with other patients and exchange their experience to reduce the burden of any social difficulties that might be provoked by asthma.
- The families of the patients will be better informed about the condition of their loved ones and will be able to secure accurate information about asthma and the proper use of medication.
- Clinicians and practitioners will be equipped with new tools and components that will provide a detailed and accurate picture of the patient’s condition outside the clinic. Analytical tools will support their decisions based on the estimates of every patient asthma evolution.
Doctors experienced coupled with automated analysis tools, will provide the possibility to provide real time feedback based on the current condition of the patient.
-Pharmaceutical companies will directly evaluate their new medicines using the modelling approaches developed under myAirCoach project.
- Healthcare systems will see the cost of asthma reduced in the long term, based on the optimization of treatments, the increased knowledge of patients and the adherence to their medication schedule.
By Whom
MyAirCoach consortium is composed by 3 clinical partners specialized in respiratory medicine and asthma treatment, 3 ICT research and academic organizations of scientific excellence, 3 medical private manufacturers focused on improving treatments for patients with inflamed airways, 2 asthma and respiratory diseases
patients’ associations and 1 respiratory and allergy pharmaceutical company.
For more information please visit the project’s website at: myAirCoach.eu

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

"1 Analysis of current practices and state of the art
As a first step towards the understanding of the commercial and research environment of asthma mHealth solutions a detailed review of the available products and approaches was conducted within the first months of the project. The performed analysis covered three fundamental pillars of asthma self-management, namely: 1) inhaler devices and sensors for asthma’s disease, 2) computational modelling and prediction of lung behavior and 3) self-support management and personal guidance systems.
Related Submitted Deliverables: D1.1 “Analysis of current practices”
Related Journal Publication: D. Kikidis, K.Votis, D. Tzovaras, O. Usmani. "The Digital Asthma Patient: History and Future of Inhaler Based Health Monitoring Devices". To be published in the Journal of Aerosol Medicine and Pulmonary Drug Delivery (2016)

2 Assessment and analysis of user requirements
The collection of user needs and requirements is the fundamental basis for the development of any system or product. Furthermore, the accurate understanding of user requirements and is specifically important for the introduction of solutions in the healthcare environment since they can help to protect the safety and privacy of users. In this direction the MyAirCoach project involved the participation of 249 users (187 patients and 62 health care professionals) and collected their feedback regarding important issues of the proposed solution and the deployment of the project’s workplan. The collected responses were analyzed and significant results were presented in detail together with the procedures followed and the measures taken for the protection of the privacy of all the participants.
Related Submitted Deliverables: D1.2 “Use requirements, use cases, UCD methodology and final protocols for evaluation studies”

3 Definition of use case scenarios to be used in the validation processes
A fundamental component of the project’s work plan is the development of the validation campaigns which will be used as ultimate and objective assessment of the MyAirCoach system (see Project Objective 7). In order to guarantee a proper deployment of the related evaluation tasks (see Work-Package 6) and also provide a number of important targeted indicators of system usability and functionality, the collected user requirements were combined with the project’s DoA for the definition of 24 use case scenarios separated into four main categories based on their primary user groups, namely: 1) Exclusive patient oriented use cases, 2) Use cases of families and caregivers of patients, 3) Health care professional oriented use cases and finally 4) Use cases covering medical research activities
Related Submitted Deliverables: D1.2 “Use requirements, use cases, UCD methodology and final protocols for evaluation studies”

4 Planning or the User Centered Design processes of the project
The design and development processes of the MyAirCoach project will be based on user centered approaches that will put the need and requirements of all user groups in the center of the final system towards its increased usability and usefulness. In this direction a detailed User Centered Design plan was developed under the work of T1.2 of the project and will form the basis for the involvement of patient and their families as well as doctors and medical researchers in the following stages of the project. More specifically, the MyAirCoach UCD plan included the definition a model of UCD based on the ISO 13407 standard enhanced with characteristics of the waterfall model of system development. As the next step the user assessment methods were separated in three fundamental categories and they were analyzed in regards to their deployment, results, advantages/disadvantages as well as their expected cost and optimal number of participants. In this way a reference manual was formed for the use of these methods in the timeline of the project. Furthermore, the second part of the UCD plan included the discussion of the above described approaches in regards to the specific work packages and tasks of the project identifying possible areas of user involvement and outlining import first steps in this direction. Finally, the Advisory Patient Forum of the project was discussed as the main tool for the collection of feedback by the patient community based on the online platform that was developed within the first six months of the project.
Related Submitted Deliverables: D1.2 “Use requirements, use cases, UCD methodology and final protocols for evaluation studies”. D6.1 “Assembly of the consultative patient forum”, D7.2 “Project web presence (website, wiki, blog, social media)”

5 List of identified technical requirements of the MyAirCoach system
A crucial step towards the translation of the user requirements and the defined use case scenarios into a functional and useful system is the definition of a set of fundamental technical requirements that should be addressed by the project. In this direction the work of MyAirCoach in the first reporting period included a detailed list of technical specifications as they relate to the expected functionalities of final system. More specifically, the list of technical specifications covered both functional and non-functional requirements which were further categorized based on their connection with specific tasks of the project and their positioning in MyAirCoach architecture. Each one of the identified requirements was described in detail and included the respective measurable fit criterion. Furthermore, all requirements were prioritized in terms of usefulness and technical difficulty and were assigned to initially selected consortium partners. In order to facilitate the introduction of new technical requirements and the revision of the current list, a specification framework was defined based on the Volere method.
Related Submitted Deliverables: D1.3 “MyAirCoach technical specifications and end-to-end architecture (first version)”

6 Definition of the first version of the system’s architecture
The architecture of a software system is maybe the fundamental basis upon which all design and development processes should be based. In this direction the work of the MyAirCoach in the first reporting period included the definition of the first version of the system’s architecture, putting special focus on the efficient integration of component in the following stages of the project and as they are planed within T5.4 “System Integration”. More specifically, the different viewpoints of the architecture were described and the methodology of their definition was outlined. As the following step the conceptual architecture of MyAirCoach, as it was presented in the Annex of the project, was enhanced and adapted to the identified user and technical requirements. The proposed first version of the architecture was then described as it has been separated into four main layers: the knowledge layer, the system layer, the application layer and the sensors layer. After a summary of the functional modules of every layer and their interactions, the envisioned interfaces were describes in detail. UML diagrams were created for every module for a more intuitive and concise representation of its functionalities and interfaces.
Related Submitted Deliverables: D1.3 “MyAirCoach technical specifications and end-to-end architecture (first version)”

7 Definition of the test campaign methodology
The aim of the myAirCoach test/quantification campaigns is to support the development of a system that will assist patients in recognizing loss of asthma control and the onset of an exacerbation and to timely adjust behaviors and/or treatments in a timely manner. With this regards, the deployment of test campaigns was planned taking into consideration both ethic and safety issues, as well as the overall scheduling of tasks and procedures. In addition, the proposed plan included a concise overview of the sensing devices and of the measurements that may potentially be included in the first quantification campaign. The final selection of methodologies was based on a careful validation of potentially interesting parameters with respect to asthma self-management and in regards to the technologies that are available within and outside the myAirCoach consortium. Furthermore, the proposed plan was designed to be balanced between the scientific wish-list of data on the one hand and the feasibility of measurements for patients and costs on the other. At last, the detailed research protocols were submitted to and approved by the responsible ethics committees in the UK and Netherlands. The definitive methodology of the second quantification campaign will be established in 2016, since this will be partly based on the experience and results of the first campaign and on the results of other work-packages.
Related Submitted Deliverables: D2.1 “Test campaign methodology”

8 Definition of the ethical plan of the project and compliance with institutional ethical requirements
In the first months of the project a detailed ethical and safety manual was created for MyAirCoach in order to serve as reference for all the work plan activities including the participation of patients. Furthermore, the deployment of assessment campaigns carried out within Task 1.2, as well as the planning for the test/quantification campaigns of Work Package 2 have been carefully designed and approved by all the responsible ethical committees in the UK and The Netherlands.
Related Submitted Deliverables: D1.2 “Use requirements, use cases, UCD methodology and final protocols for evaluation studies”. D2.1 “Test campaign methodology”, D8.3 “Ethics, safety and mHealth barriers manual”

9 Definition of the sensor components of the MyAirCoach monitoring devices
In order to design the sensing infrastructure of the MyAirCoach project the development tasks of the project should balance between technical feasibility, clinical significance and user acceptance. In this direction and under the current task, the data flow of the MyAirCoach sensing infrastructure was introduced with regard to a novel wearable mHealth system, A detailed review of sensing modalities was presented by considering the feedback of both technical and clinical partners of the consortium. The proposed wearable set of sensors is assembled of several devices that are wirelessly connected to the smartphone via Bluetooth links. Amongst other features, these devices are aiming to monitor usage statistics of the asthma inhaler, biomarkers in exhaled breath, the patient’s physiological data as well as environmental data of the patient’s surrounding. Finally a pre-selection of suitable sensors for required assembly of novel monitoring hardware prototypes has been also presented together with the communication infrastructure for the myAirCoach system, as it will be based on the LinkWatch open monitoring platform including respective interfaces and services.
Related Submitted Deliverables: D3.1 “Definition of the sensor components and the communication strategy”

10 Hardware and software design of the MyAirCoach monitoring devices
The implementation stage of the MyAirCoach Body Area Network and the laboratory testing of the envisioned devices and sensing modalities have been presented in a detailed report.
Extensive tests in laboratory sensor settings have proven that the principle of using accelerometer sensors for monitoring usage of the smart inhaler prototype is feasible. Reading data from the 9-Degrees-of-Freedom sensor, which is built into the inhaler add-on as BAN component, renders it possible to assess whether the inhaler has been shaken and the actual orientation of the inhaler. This sensing modality is expected to be used for the creation of the algorithmic components that will allow the accurate understanding of the inhaler’s actuation technique used by patients.
Another investigated sensing component is the module for the assessment of pollution in terms of particle concentrations (PM2.5 and PM10), density of NO2 and SO2 as it is envisioned to be built around a common spacer. This modality is called biomonitor, as it enables to verify potential asthma triggers in the direct environment of the patient. Moreover, the biomonitor will integrate an acoustic sensor for investigating assessment breath sounds and respiratory frequency.
Finally, breath temperature sensing capabilities were investigated. Common digital temperature sensors proved to be unsuitable for this purpose, as they cannot respond fast enough so as to measure the changes in temperature as they are caused by exhalation. In order to solve this issue the usage of an infrared sensor has been investigated with promising results in laboratory settings.
Related Submitted Deliverables: D3.2 “Final design of the hardware and embedded software”

11 Definition of the patient modelling and representation framework
In the modern health care environment the digital representation of the patient records is considered of crucial importance as it is promising the integration of novel technologies and the use of innovative algorithmic approaches for the support of not only health care professionals but also patients themselves. Unfortunately, modern Electronic Health Record representations fail to include environmental parameters as well as parameters related to the patients’ lifestyle such as activity levels and nutritional habits. In this direction the MyAirCoach project has took the first step for the formation of a patient modelling framework that will allow the collection, processing and presentation of such information based on the envisioned Body Area Network. Furthermore, the proposed framework was designed to include parameters for the computational modelling of the lungs which will support medical research and allow the enrichment of the MyAirCoach framework with the simulation of breath within the lungs.
Related Submitted Deliverables: D4.1 “Patient modelling and representation framework”

12 Signal processing approaches for the detection of inhaler actuations
One of the most important and difficult problems of effective asthma management is the decreased levels of patient adherence to the prescribed medication regiment. Usually, patients forget their medication doses or lose interest as the symptoms of asthma start to decrease. Unfortunately, this behavior has significant long term negative effects on their health. In order to address this issue a number of devices have been introduced based on the mechanical detection of inhaler activations (simple electrical buttons on the inhaler canister). Unfortunately these approaches cannot allow the understanding or whether the patient used the medication in a correct manner, e.g proper timing of breathing in and pushing button, holding breath for an adequate amount of time, etc.
In order to address these issues the MyAirCoach project is aiming to utilize the use of audio signals for the detection of inhaler actuations and the accurate assessment of the proper inhaler technique. As a first step in this direction a variety of algorithms have been tested for the detection of inhaler actuations including: 1) Logistic regression with LASSO shrinkage, 2) random forest, 3) AdaBoost, 4) Support Vector Machines , 5) Naïve-Bayes Classifier and 6) Convolution Neural Networks. The data used for the training and testing of these algorithms were collected during the activation of inhalers in the open air and not during the actual use by patients. The final results of all approaches show results of high accuracy and support in this way the extension of the algorithmic approaches to the problem of technique assessment in addition to the detection of actuations.

13 Computational modelling of the airways and respiratory system
The employment of computational mo
delling on airways and pulmonary system to assess critical parameters of asthma exacerbations summarizes the objective of MyAirCoach computational module. To pursue this goal the concepts of Computational Fluid Dynamics (CFD) and Fluid Particle Tracing (FPT) are utilized on the analysis of the behavior of the human respiratory system. In particular, Finite Element Method (FEM) is adopted to calculate the parameters of interest, such as the air velocity and pressure on the airways volume, along with particles deposition. This approach involves important steps such as geometry reconstruction, input of environmental and physiological factors, analysis, post-processing and visualization of the results. Specifically, the geometry obtained from CT scans is delivered as a surface mesh. Opensource software, such as Blender and Meshlab, are used to pre-process the geometry (i.e. surface reconstruction, remeshing, triangulation, and decimation) in order to provide smoothness and facilitate the upcoming procedures of meshing and solving. Environmental and physiological factors are taken into account. Temperature, pressure, and particle properties, such as particle density, mass flow and particle mass, are needed as input of the CFD and FPT models. The implementation of the FEM algorithm dictates specific steps to be followed, i.e. volume meshing, application of boundary conditions and solving. The SnappyHexMesh algorithm is used to provide a detailed volume mesh of the airways considering the snapping procedure to the surface mesh along with appropriate boundary layers. The next step requires applying suitable boundary conditions on the surface mesh. Important variables such as velocity, pressure and kinematic viscosity are initialized depending on the physical parameters of respiration. Additionally, open source software, such as OpenFoam, is used to accurately solve for the output parameters of the FEM model. Turbulence models such as the SST k-omega model are adopted to extract the detailed characteristics of the airflow. Particularly, air velocity and pressure are calculated on the entire airways volume. Post-processing of the results and visualization are performed via Paraview, an opensource tool dedicated on this purpose. The data acquired from the CFD analysis are used as initial conditions to apply FPT models to investigate the movement and deposition of pollutant, allergen and drug particles. In this direction, an appropriate mapping of the previously generated mesh and the necessary fields is implemented, whereas suitable boundary conditions are applied. Future work involves solving of the particles movement models to extract important quantities such as particle trajectories and deposition on the airways.

14 MyAirCoach education and training approaches for the support of patients
An important component for the proper management of asthma by patients themselves is their knowledge regarding the pathophysiology of the disease, its symptoms and how medication affects the overall health of the respiratory system. Unfortunately, modern approaches utilize static presentation approaches in the form of document instructions or informative videos. MyAirCoach has taken one step forward for the development of a 3D interactive environment where patients can see how to properly utilize their inhaler. Furthermore a mobile application environment was designed that is enriched with multiple choice functionalities that support the engagement of patients in understanding their disease and promote asthma knowledge through the use of gamification approaches.

15 Development of the initial version of the online doctor environments
A first version of the online doctor environment was developed that supports the presentation of patient history and the grouping of assessments and exams based on their time and chronological order. Furthermore the proposed environment is designed to allow the incorporation of educational components as presented in the previous section. It is important to underline in this point that the suboptimal use of inhalers is not only a matter of patients but is also related to reduced knowledge and experience of medical professionals also.

16 Assembly of the Advisory Patient Forum
As already mentioned in the previous sections, MyAirCoach will be based on User Centered Design approach which is expected to strengthen the usability of the final system components and direct the goals of the project in areas that have a direct and significant positive impact in the patients’ life. In this direction the MyAirCoach Advisory Patient Forum was assembled in the first six months of the project. The APF is formed by 22 European asthma patients who have been contributed to project activities and deliverables. Furthermore, an online platform was created for the fast and easy communication of participants with the project’s consortium.
Related Submitted Deliverables: D6.1 “Assembly of the consultative patient forum”

17 Horizontal activities for the dissemination of the project’s activities
The dissemination of the project results is considered of fundamental importance for not only the support of possible commercial exploitation of developed components but also for the creation of strong links of the project outcomes with society and the support of patients, their families and the health care environment as a whole. In this direction increased effort has been given for the dissemination of the project results and the participation of the consortium in international meetings and conferences. A visual identity was developed to brand the project and a MyAirCoach brochure was produced to promote and create awareness on the project. Furthermore, the project’s website together with accounts in the most popular social media have been created and maintained from the first months of the project’s timeline.
Related Submitted Deliverables: D7.1 “Dissemination plan and MyAirCoach dissemination material”, D7.2 “Project web presence (Website, Wiki, Blog, Social Media)”, D8.2 “Periodic management report for M6”, D8.3 “First year project periodic report”

18 Definition of the project’s Data Management Plan
A detailed data management plan was developed so as to support the life cycle for all data that will be collected, processed or generated by the project. The data management plan of the project consists of a detailed analysis of the datasets that the partners of the MyAirCoach project plan to collect and use. Foreseen datasets contain inhaler usage measurements, physiology assessments, exhaled Nitric Oxide measurements, environmental measurements, patient tomography data, virtual models etc. Each dataset was separately analyzed, with emphasis given on the nature of the data, the accessibility and its possible access type, as well as any ethical issues that may arise from manipulating sensitive personal information. Furthermore, the open data repository of the project was demonstrated as it will be integrated with the online MyAirCoach platform and allow the access of the project outcomes and datasets for the stimulation of research in the long term extending the 3 year timeline of the project’s workplan
Related Submitted Deliverables: D7.6 "Data Management Plan"

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

"1 Architecture of a mHealth system for the support of asthma self-management
MyAirCoach proposes an innovative approach not only as a concept of asthma self-management but also as the architecture of the developed solution in terms of the integration of a variety of sensing components in a Body Area Network and the utilization of the collected data for the personalization of decision support and the asthma management functionalities through modeling approaches. This framework is aiming to optimize asthma management improving in this way the patients’ quality of life and also increase the efficiency of the healthcare system as a whole through the reduction of patient visits in the healthcare environment (scheduled and emergency) through the online supervision of patients by their doctors and the parallel support of their families and social environment
Related Conference Publication: K. Votis, A. Lalos, K. Moustakas, D. Tzovaras. "Analysis, Modelling and Sensing of both Physiological and Environmental Factors for the Customized and Predictive Self-Management of Asthma". In proceedings of the 6th Panhellenic Conference of Biomedical Technology. Athens, Greece, 6-8 May 2015.

2 Signal processing of the detection of inhaler actuations
Although neural networks have been utilized for a variety of classification problems a very small number of research studies have focused on the biomedical applications that utilize acoustic data. In this direction MyAirCoach proposes a novel application of Convolution Neural networks in the area of sound processing and used the proposed approach for the detection of inhaler actuations in the natural living environment by performing experiments on data collected outside the confined environment of the laboratory. Furthermore, the proposed methodology takes into account the protection of the privacy of users and utilizes a subsection of the vocal frequency spectrum. Finally, the proposed methodology will be the basis for the next steps of monitoring the overall inhaler technique as it comprises of the timing of inhalations and exhalations and the positioning of the inhaler actuation relative to the inhalation of the users.
Related Conference Publication: D. Kikidis, K. Votis, D. Tzovaras. "Utilizing Convolution Neural Networks for the Acoustic Detection of Inhaler Actuations". In Proceedings of the IEEE International Conference on E-Health and Bioengineering. Iasi, Romania, 19-21 November 2015.

3 Patient modelling and representation framework
As already mentioned, the MyAirCoach solution proposes an innovative framework for the modelling of patients that includes environmental and lifestyle parameters in addition to the commonly covered clinical parameters.
Related Publication: K.Votis, D. Kikidis, D. Tzovaras, O. Usmani. "The Digital Patient: The Future of Mobile Health for Respiratory Patients". In proceedings of the 2015 Congress of the International Society for Aerosols in Medicine. Munich, Germany, 30 May- 3 June 2015.

4 Computational modeling of respiratory system
There are several studies related to computational modeling of different aspects of the pulmonary function. These include aspects related to airflow analysis and ventilation, particle transportation and deposition and mechanical properties of the lung airways tissue. However, their major drawback is that they refer to healthy lungs without taking into account structural alterations present in obstructive lung diseases. In addition to the aforementioned models there are applications that focus on the lung geometry allowing either to reconstruct the 3D lung model from CT scans or to view and study the lung geometry without proving the ability to perform alterations appropriate for applying computational fluid dynamics simulation. Thus, their major drawback is that they do not combine the ability to simulate geometry alterations present in obstructive lung diseases with computational fluid dynamics simulations.
To cope with the aforementioned issues and allow the processing of patient specific lung airways geometry we propose a method that combines the deformation of the geometry in a manner that simulates bronchoconstriction. More specifically, the proposed approaches are executed directly on already available 3D models of the lung, extracted from CT scans, by using conventional SoA methods. The user interface allow the user to select the area of interest, e.g either a specific branch, or all the branches of the i-th generation, or the whole model itself and perform narrowing by executing a Laplacian mesh contraction approach. The surface of the given object is iteratively contracted either in the direction of the inward normal or according to a customized function that deforms the mesh in a user selected manner. A skeleton extraction technique has been also developed for converting the 3D object into a 1D curve skeleton that is essential for airway segmentation, for personalization and for predicting the structure of airways in models that include more than nine generations of branching.
The developed toolset allows the view and deformation of lung 3D structures in order to be used for computational fluid dynamics, allowing the visualization of the airflow for different kinds of geometry. The output of the CFD simulation is essential for predicting particle deposition upon the inner part of the airway walls, allowing: i) the clinician to study the way the drug or other harmful particles that cause inflammation are dispersed inside the lungs for different stages of a crisis and for different levels of inflammation ii) the user to determine the effectiveness of a delivery system upon inflamed airways and use the results as input for assessing which parts of the patient’s lung are more easily affected and predicting an obstruction of a specific airway.
Related Conference Publication: S. Nousias, A.S. Lalos, K. Moustakas, "Computational Modeling for Simulating Obstructive Lung Diseases based on Geometry Processing Methods", To be presented HCI International 2016, Toronto, Canada, 17 - 22 July 2016.

5 Education and training components for effective asthma management
Efficient patient education is considered a fundamental component in this process as it holds the promise to positively affect all the above aspects of asthma self-management. In this direction the MyAirCoach mobile application was designed to include an education framework that aims to increase the awareness of patients about the disease of asthma and help them improve their technique of inhaler use through illustrative presentations and interactive multiple choice questions. The current paper outlines the development process and selected structure of this environment, and takes a first step towards its validation through the first round of feedback collected from the MyAirCoach Advisory Patient Forum.
Related Conference Publications: D. Kikidis, K. Votis, D. Tzovaras. "MyAirCoach: Designing a Mobile Application for the Education of Patients Regarding Asthma Disease". In proceedings of the international Conference on Interactive Mobile Communication, Technologies and Learning. Thessaloniki, Greece, 19-20 November 2015.

6 Energy efficient transmission of sound measurements related to respiratory health
Modern wearables in general and health monitoring devices in particular are based on the continuous communication with online resources and the transmission of large amounts of data in their long term use. This transmission of information introduces an great burden in the power consumption of devices reducing significantly the battery life of modern mobile and wearable devices. In this direction two novel compression approaches have been proposed for increased transmission efficiency of sound segments of patient breath and 3D representations of the lung airways.
For the case of audio compressed sensing the proposed solution focuses on the enhancement of the state of the art by taking into account specific generic characteristics (e.g., block sparsity, sample correlation) of the breath sounds during their reconstruction at the Body Node Coordinator. This is achieved by applying a novel recovery algorithm; named PCA based Group LASSO, that exploits the block sparsity of the signal eigen-spectrum, increasing significantly the sensing system energy efficiency as compared to traditional CS recovery schemes.
Related Publications: A. Lalos, K. Moustakas. "Energy Efficient Telemonitoring of Asthmatic Wheezes". In proceedings of the 2015 European Signal Processing Conference. Nice, France, 31 August- 4 September 2015.

7 Energy efficient transmission of models of lung geometry
With the growing demand for easy and reliable generation of 3D models representing real-world objects and environments in mobile cloud computing platforms, new schemes for acquisition, storage and transmission of 3D meshes are required. In general, 3D meshes consist of two distinct components: vertex positions and vertex connectivity. Vertex position encoders are much more resource demanding than connectivity encoders, stressing the need for novel geometry compression schemes. The proposed compression/reconstruction approaches minimize the samples that are required for transmission yet assuring accurate reconstruction at the receiver, by exploiting specific local characteristics of the surface geometry in the graph Fourier domain. Simulation studies show that the proposed schemes, as compared to the state of the art approaches, achieve competitive Compression Ratios (CRs), offering at the same time significantly lower compression computational complexity, which is essential for mobile cloud computing platforms. The proposed approach was applied in the modelling physiological information of lung geometry.
Related Publications: A. Lalos, I. Nikolas, K. Moustakas. "Sparse Coding of Dense 3D Meshes in Mobile Cloud Applications". In proceedings of IEEE International Symposium on Signal Processing and Information Technology. Abu Dhabi, United Arab Emirates, 7-10 December 2015. Received the Best Paper Award"

Related information

Record Number: 190371 / Last updated on: 2016-11-15
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top