Remote Monitoring and Automatic Fall Detection for Elderly People at Home

Mälardalen University Press Licentiate Theses No. 207

REMOTE MONITORING AND AUTOMATIC FALL

DETECTION FOR ELDERLY PEOPLE AT HOME

Gregory Koshmak

2015

School of Innovation, Design and Engineering

Mälardalen University Press Licentiate Theses

No. 207

REMOTE MONITORING AND AUTOMATIC FALL

DETECTION FOR ELDERLY PEOPLE AT HOME

Gregory Koshmak

2015

Copyright © Gregory Koshmak, 2015 ISBN 978-91-7485-212-7

ISSN 1651-9256

Printed by Arkitektkopia, Västerås, Sweden

Abstract

Aging population is a one of the key problems for the vast majority of so called ”more economically developed countries” (MEDC). The amount of elderly people who suffer from multiple disease and require permanent monitoring of their vital parameters has increased recently resulting in extra healthcare costs. Modern healthcare systems exploited in geri-atric medicine are often obtrusive and require patients presence at the hospital which interferes with their demand in independent life style. Recent developments on telecare market provide a wide range of wire-less solutions for distant monitoring of medical parameters and health assistance. However, most of the devices are programmed for spot check-ing and operate independently from each other. There is still a lack of integrated framework with high interoperability and on-line continuous monitoring support for further correlation analyses. The current study is a step towards complete and continuous data collection system for el-derly people with various types of health problems. Research initiative is motivated by recent demand in reliable multi-functional remote monitor-ing systems, combinmonitor-ing different data sources. The main focus is made on fall detection methods, interoperability, real-life testing and correla-tion analyses. The list of main contribucorrela-tions contains (1) investigating communication functionalities, (2) developing algorithm for reliable fall detection, (3) multi-sensor fusion analyses and overview of the latest multi-sensor fusion approaches, (4) user study involving healthy volun-teers and elderly people. Evaluation is performed through a series of computer simulation and real-life testing in collaboration with the local medical authorities. As a result we expect to obtain a monitoring sys-tem with reliable communication capabilities, inbuilt on-line processing, alarm generating techniques and complete functionality for integration with similar systems or smart-home environment.

Abstract

Aging population is a one of the key problems for the vast majority of so called ”more economically developed countries” (MEDC). The amount of elderly people who suffer from multiple disease and require permanent monitoring of their vital parameters has increased recently resulting in extra healthcare costs. Modern healthcare systems exploited in geri-atric medicine are often obtrusive and require patients presence at the hospital which interferes with their demand in independent life style. Recent developments on telecare market provide a wide range of wire-less solutions for distant monitoring of medical parameters and health assistance. However, most of the devices are programmed for spot check-ing and operate independently from each other. There is still a lack of integrated framework with high interoperability and on-line continuous monitoring support for further correlation analyses. The current study is a step towards complete and continuous data collection system for el-derly people with various types of health problems. Research initiative is motivated by recent demand in reliable multi-functional remote monitor-ing systems, combinmonitor-ing different data sources. The main focus is made on fall detection methods, interoperability, real-life testing and correla-tion analyses. The list of main contribucorrela-tions contains (1) investigating communication functionalities, (2) developing algorithm for reliable fall detection, (3) multi-sensor fusion analyses and overview of the latest multi-sensor fusion approaches, (4) user study involving healthy volun-teers and elderly people. Evaluation is performed through a series of computer simulation and real-life testing in collaboration with the local medical authorities. As a result we expect to obtain a monitoring sys-tem with reliable communication capabilities, inbuilt on-line processing, alarm generating techniques and complete functionality for integration with similar systems or smart-home environment.

Sammanfattning

En ˚aldrande befolkning utg¨or ett av de viktigaste problemen f¨or de allra flesta s˚a kallade ”mer ekonomiskt utvecklade l¨ander” (MEDC). M¨angden ¨

aldre m¨anniskor som lider av multi-sjukdomar och kr¨aver st¨andig ¨overvakning av vitala parametrar har ¨okat p˚a senare tid, vilket resulterar i ¨okade sjukv˚ardskostnader. Geriatrikens moderna sjukv˚ardssystem kr¨aver ofta att patienterna ¨ar n¨arvarande p˚a sjukhuset, vilket kraftigt begr¨ansar en sj¨alvst¨andig och oberoende livsstil. Den senaste utvecklingen p˚a telemedicinomr˚adet erbjuder ett brett utbud av tr˚adl¨osa l¨osningar inom h¨alsov˚ard f¨or distans¨overvakning av medicinska parametrar. De flesta l¨osningarna inneb¨ar punktkontroll av enskilda parametrar och arbetar oberoende av varandra. Det saknas fortfarande integrerade l¨osningar med h¨og interoperabilitet och kontinuerlig on-line ¨overvakningsst¨od f¨or att kunna genomf¨ora ytterligare korrelationsanalyser. Detta arbete utg¨or ett steg mot ett fullst¨andigt och kontinuerligt datainsamlingssystem f¨or ¨aldre personer med olika typer av h¨alsoproblem. Forskningsinitiativet motiveras av senaste tidens efterfr˚agan p˚a tillf¨orlitliga multifunktionella system f¨or distans¨overvakning, som kombinerar olika datak¨allor. Huvud-fokus utg¨ors av falldetektionsmetoder, interoperabilitet, verkliga tester och korrelationsanalyser. Listan ¨over de fr¨amsta bidragen inneh˚aller (1) att unders¨oka kommunikationsfunktionaliteter, (2) utveckla en al-goritm f¨or tillf¨orlitlig falldetektion, (3) multisensor-fusion-analyser och ¨oversikt ¨over multisensor-fusion-strategier, (4) en anv¨andarstudie med friska frivilliga ¨aldre. Utv¨arderingen sker genom en serie av datorsimu-leringar och tester i verklig milj¨o i samarbete med lokala h¨also- och sjukv˚ardsmyndigheter. M˚alet ¨ar ett ¨overvakningssystem med tillf¨orlitliga kommunikationsm¨ojligheter, inbyggd on-line-bearbetning, tekniker f¨or larmgenerering och funktionalitet f¨or integration med liknande system eller i en smart hemmilj¨o.

Sammanfattning

En ˚aldrande befolkning utg¨or ett av de viktigaste problemen f¨or de allra flesta s˚a kallade ”mer ekonomiskt utvecklade l¨ander” (MEDC). M¨angden ¨

aldre m¨anniskor som lider av multi-sjukdomar och kr¨aver st¨andig ¨overvakning av vitala parametrar har ¨okat p˚a senare tid, vilket resulterar i ¨okade sjukv˚ardskostnader. Geriatrikens moderna sjukv˚ardssystem kr¨aver ofta att patienterna ¨ar n¨arvarande p˚a sjukhuset, vilket kraftigt begr¨ansar en sj¨alvst¨andig och oberoende livsstil. Den senaste utvecklingen p˚a telemedicinomr˚adet erbjuder ett brett utbud av tr˚adl¨osa l¨osningar inom h¨alsov˚ard f¨or distans¨overvakning av medicinska parametrar. De flesta l¨osningarna inneb¨ar punktkontroll av enskilda parametrar och arbetar oberoende av varandra. Det saknas fortfarande integrerade l¨osningar med h¨og interoperabilitet och kontinuerlig on-line ¨overvakningsst¨od f¨or att kunna genomf¨ora ytterligare korrelationsanalyser. Detta arbete utg¨or ett steg mot ett fullst¨andigt och kontinuerligt datainsamlingssystem f¨or ¨aldre personer med olika typer av h¨alsoproblem. Forskningsinitiativet motiveras av senaste tidens efterfr˚agan p˚a tillf¨orlitliga multifunktionella system f¨or distans¨overvakning, som kombinerar olika datak¨allor. Huvud-fokus utg¨ors av falldetektionsmetoder, interoperabilitet, verkliga tester och korrelationsanalyser. Listan ¨over de fr¨amsta bidragen inneh˚aller (1) att unders¨oka kommunikationsfunktionaliteter, (2) utveckla en al-goritm f¨or tillf¨orlitlig falldetektion, (3) multisensor-fusion-analyser och ¨oversikt ¨over multisensor-fusion-strategier, (4) en anv¨andarstudie med friska frivilliga ¨aldre. Utv¨arderingen sker genom en serie av datorsimu-leringar och tester i verklig milj¨o i samarbete med lokala h¨also- och sjukv˚ardsmyndigheter. M˚alet ¨ar ett ¨overvakningssystem med tillf¨orlitliga kommunikationsm¨ojligheter, inbyggd on-line-bearbetning, tekniker f¨or larmgenerering och funktionalitet f¨or integration med liknande system eller i en smart hemmilj¨o.

Acknowledgement

First and foremost let me offer my sincerest gratitude to supervisors Maria Linden, Amy Loutfi and Matts Bj¨orkman, who have constantly been a great inspiration for me. From the very moment I have started my academic carrier I always know where to go in times of doubt and desperation. Your expert knowledge, academic experience and wise ad-vice is guiding me through the dark forest of research life. I am lucky to work with such professional researches and extremely pleasant people.

I have not been spending in V¨aster˚as every day during the past years, nut I can honestly enjoy IDT team and appreciate all the moments we shared. I would like to personally thank my ”old” office room mates Nikola, Marcus, Jimmy, Martin and my current desk neighbors Sara, Per, Arash and . It is always nice to be surrounded by intelligent and at the same time cheerful people with a great sense of humor. Thanks for being brave to participate in my experiments and thanks for challenging me with intellectual discussions. I can not help mentioning our annual journeys with IFT group, this was something I will remember for a long time!

Thanks to all the members of the GiraffPlus project, especially Fil-ippo, Ales, Jonas, , with whom I was collaborating a lot during the project lifetime. Additional words of gratitude go to Stig and Uno, who in spite of their advanced age have charged me with unconditional en-ergy and positive attitude towards life.

A huge thanks to my family. There are no such words to describe how much they all mean to me. My parents and grandparents have always been a source of unlimited inspiration and strong support. Research work can be tough sometimes, but it is not even close to what they had to go through during the last 1.5 yeah. And yet they always managed to find some encouraging words, wise advice and just a funny joke to share.

Acknowledgement

First and foremost let me offer my sincerest gratitude to supervisors Maria Linden, Amy Loutfi and Matts Bj¨orkman, who have constantly been a great inspiration for me. From the very moment I have started my academic carrier I always know where to go in times of doubt and desperation. Your expert knowledge, academic experience and wise ad-vice is guiding me through the dark forest of research life. I am lucky to work with such professional researches and extremely pleasant people.

I have not been spending in V¨aster˚as every day during the past years, nut I can honestly enjoy IDT team and appreciate all the moments we shared. I would like to personally thank my ”old” office room mates Nikola, Marcus, Jimmy, Martin and my current desk neighbors Sara, Per, Arash and . It is always nice to be surrounded by intelligent and at the same time cheerful people with a great sense of humor. Thanks for being brave to participate in my experiments and thanks for challenging me with intellectual discussions. I can not help mentioning our annual journeys with IFT group, this was something I will remember for a long time!

Thanks to all the members of the GiraffPlus project, especially Fil-ippo, Ales, Jonas, , with whom I was collaborating a lot during the project lifetime. Additional words of gratitude go to Stig and Uno, who in spite of their advanced age have charged me with unconditional en-ergy and positive attitude towards life.

A huge thanks to my family. There are no such words to describe how much they all mean to me. My parents and grandparents have always been a source of unlimited inspiration and strong support. Research work can be tough sometimes, but it is not even close to what they had to go through during the last 1.5 yeah. And yet they always managed to find some encouraging words, wise advice and just a funny joke to share.

List of Publications

Papers included in the licentiate thesis

1

Paper A Evaluation of the Android-Based Fall Detection System with

Phys-iological Data Monitoring. Gregory Koshmak, Maria Linden, Amy

Loutfi, 35th Annual International Conference of the IEEE EMBS Osaka, Japan, July 3-7, 2013.

Paper B Dynamic Bayesian Networks for Context-Aware Fall Risk

Assess-ment. Gregory Koshmak, Maria Linden, Amy Loutfi, A Special

Is-sue of Ambient Assisted Living (AAL): Sensor, Architectures and Applications.

Paper C Challenges and Issues in Multi-Sensor Fusion Approach for Fall

Detection: Review Paper. Gregory Koshmak, Maria Linden, Amy

Loutfi. Submitted to Journal of Sensors.

1_{The included articles have been reformatted to comply with the licentiate layout}

List of Publications

Papers included in the licentiate thesis

1

Paper A Evaluation of the Android-Based Fall Detection System with

Phys-iological Data Monitoring. Gregory Koshmak, Maria Linden, Amy

Loutfi, 35th Annual International Conference of the IEEE EMBS Osaka, Japan, July 3-7, 2013.

Paper B Dynamic Bayesian Networks for Context-Aware Fall Risk

Assess-ment. Gregory Koshmak, Maria Linden, Amy Loutfi, A Special

Is-sue of Ambient Assisted Living (AAL): Sensor, Architectures and Applications.

Paper C Challenges and Issues in Multi-Sensor Fusion Approach for Fall

Detection: Review Paper. Gregory Koshmak, Maria Linden, Amy

Loutfi. Submitted to Journal of Sensors.

1_{The included articles have been reformatted to comply with the licentiate layout}

x

Additional papers, not included in the

licen-tiate thesis

Conferences

• A smart-Phone Based Monitoring System with Health Device Pro-file for Measuring Vital Physiological parameters. Gregory Koshmak,

Martin Ekstr¨om, Maria Linden. World Congress on Medical Physics and Biomedical Engineering, Beijing, China, May 26-31, 2012

• Heart Rate Measurement as a tool to quantify Sedentary Behav-ior. Anna ˚Akeberg, Gregory Koshmak, Anders Johansson, Maria Linden, International Conference on Wearable Micro and Nano Technologies for Personalized Health, V¨aster˚as, Sweden, June 2-4, 2015.

Contents

I

Thesis

1

1 Introduction 3 1.1 Motivation . . . 3 1.2 Problem description . . . 4 1.3 Research Hypothesis . . . 5 1.4 Research Questions . . . 5 1.5 Thesis Outline . . . 6 2 Wireless Monitoring 7 2.1 Wearable Sensors . . . 7 2.2 Fall Detection . . . 9 2.2.1 Fall Characterisitcs . . . 9

2.2.2 Context-Aware Fall Detection . . . 11

2.2.3 Smartphone-based Fall Detection . . . 12

2.3 Smart Home Environment . . . 13

3 Methodology 15 3.1 Research Approach . . . 15 3.2 GiraffPlus Project . . . 16 4 Research Contribution 19 4.1 Paper A . . . 20 4.2 Paper B . . . 21 4.3 Paper C . . . 22

5 Conclusions and Future Work 25 5.1 Conclusions . . . 25

5.2 Future Work . . . 26

x

Additional papers, not included in the

licen-tiate thesis

Conferences

• A smart-Phone Based Monitoring System with Health Device Pro-file for Measuring Vital Physiological parameters. Gregory Koshmak,

Martin Ekstr¨om, Maria Linden. World Congress on Medical Physics and Biomedical Engineering, Beijing, China, May 26-31, 2012

• Heart Rate Measurement as a tool to quantify Sedentary Behav-ior. Anna ˚Akeberg, Gregory Koshmak, Anders Johansson, Maria Linden, International Conference on Wearable Micro and Nano Technologies for Personalized Health, V¨aster˚as, Sweden, June 2-4, 2015.

Contents

I

Thesis

1

1 Introduction 3 1.1 Motivation . . . 3 1.2 Problem description . . . 4 1.3 Research Hypothesis . . . 5 1.4 Research Questions . . . 5 1.5 Thesis Outline . . . 6 2 Wireless Monitoring 7 2.1 Wearable Sensors . . . 7 2.2 Fall Detection . . . 9 2.2.1 Fall Characterisitcs . . . 9

2.2.2 Context-Aware Fall Detection . . . 11

2.2.3 Smartphone-based Fall Detection . . . 12

2.3 Smart Home Environment . . . 13

3 Methodology 15 3.1 Research Approach . . . 15 3.2 GiraffPlus Project . . . 16 4 Research Contribution 19 4.1 Paper A . . . 20 4.2 Paper B . . . 21 4.3 Paper C . . . 22

5 Conclusions and Future Work 25 5.1 Conclusions . . . 25

5.2 Future Work . . . 26

xii Contents

Bibliography 29

II

Included Papers

35

6 Paper A:

Evaluation of the Android-Based Fall Detection System

with Physiological Data Monitoring 37

6.1 Introduction . . . 39

6.2 Implementation . . . 40

6.2.1 Fall Detection Algorithm . . . 40

6.3 Experiments and Results . . . 42

6.4 Conclusion . . . 46

Bibliography . . . 49

7 Paper B: Dynamic Bayesian Networks for Context-Aware Fall Risk Assessment 53 7.1 Introduction . . . 55

7.2 Related Work . . . 56

7.2.1 Context-Aware Fall Detection . . . 56

7.2.2 Mobile Healthcare Integration . . . 58

7.3 Framework . . . 59

7.3.1 Mobile-Based Fall Detection System . . . 59

7.3.2 Context Recognition . . . 62

7.3.3 Dynamic Bayesian Network . . . 63

7.4 System Integration . . . 65

7.5 System Evaluation . . . 69

7.5.1 Matlab Simulated Model . . . 69

7.5.2 Demonstration Model . . . 70

7.5.3 Fall Risk Probability Estimation . . . 72

7.6 Conclusions and Future Work . . . 75

Bibliography . . . 77

8 Paper C: Challenges and Issues in MultiSensor Fusion Approach for Fall Detection: Review Paper 83 8.1 Introduction . . . 85

8.2 Fall detection . . . 86

Contents xiii 8.2.1 Fall Characteristics and Popular Approaches . . . 86

8.3 Sensor fusion in Fall Detection . . . 89

8.3.1 Context-aware sensors fusion . . . 91

8.3.2 Wearable sensors fusion . . . 94

8.3.3 Wearable/Ambient sensor fusion . . . 98

8.4 Discussion . . . 103

8.4.1 Challenges . . . 103

8.4.2 Future Trends . . . 104

8.5 Conclusion . . . 105

xii Contents

Bibliography 29

II

Included Papers

35

6 Paper A:

Evaluation of the Android-Based Fall Detection System

with Physiological Data Monitoring 37

6.1 Introduction . . . 39

6.2 Implementation . . . 40

6.2.1 Fall Detection Algorithm . . . 40

6.3 Experiments and Results . . . 42

6.4 Conclusion . . . 46

Bibliography . . . 49

7 Paper B: Dynamic Bayesian Networks for Context-Aware Fall Risk Assessment 53 7.1 Introduction . . . 55

7.2 Related Work . . . 56

7.2.1 Context-Aware Fall Detection . . . 56

7.2.2 Mobile Healthcare Integration . . . 58

7.3 Framework . . . 59

7.3.1 Mobile-Based Fall Detection System . . . 59

7.3.2 Context Recognition . . . 62

7.3.3 Dynamic Bayesian Network . . . 63

7.4 System Integration . . . 65

7.5 System Evaluation . . . 69

7.5.1 Matlab Simulated Model . . . 69

7.5.2 Demonstration Model . . . 70

7.5.3 Fall Risk Probability Estimation . . . 72

7.6 Conclusions and Future Work . . . 75

Bibliography . . . 77

8 Paper C: Challenges and Issues in MultiSensor Fusion Approach for Fall Detection: Review Paper 83 8.1 Introduction . . . 85

8.2 Fall detection . . . 86

Contents xiii 8.2.1 Fall Characteristics and Popular Approaches . . . 86

8.3 Sensor fusion in Fall Detection . . . 89

8.3.1 Context-aware sensors fusion . . . 91

8.3.2 Wearable sensors fusion . . . 94

8.3.3 Wearable/Ambient sensor fusion . . . 98

8.4 Discussion . . . 103

8.4.1 Challenges . . . 103

8.4.2 Future Trends . . . 104

8.5 Conclusion . . . 105

I

Thesis

I

Thesis

Chapter 1

Introduction

1.1

Motivation

Aging population has been one of the main concerns in most developed countries during the last decade [1]. Most elderly people suffer from wider spectrum of various diseases and more emergency situations such as fall are likely to occur [2]. As a result, they need to be urgently transported to the hospital, where they will be observed and provided with medical help if health condition is at risk. At the same time, the amount of elderly people choosing to maintain their independent lifestyles is growing rapidly, which makes it harder for medical profes-sionals to follow changes and trends in patient’s health conditions outside hospital environment. However, remote monitoring can help to prevent described scenario, significantly reduce healthcare costs and at the same time maintain patient’s independent lifestyle [3]. Therefore, there is a clear demand in reliable multi-functional remote monitoring systems for elderly people, which collect and combine different sources of med-ical data corresponding to everyday routine of the monitored patient. In many cases, different components comprising the systems are dis-integrated and operating separately from each other. However, if we combine monitoring components (e.g. sensors, actuators) into smart en-vironments, we will be able to carry out observations for people with multiple chronic conditions at home. It will help to improve elderly pa-tient’s level of freedom and safety, which is one of the main issues in healthcare industry.

Chapter 1

Introduction

1.1

Motivation

Aging population has been one of the main concerns in most developed countries during the last decade [1]. Most elderly people suffer from wider spectrum of various diseases and more emergency situations such as fall are likely to occur [2]. As a result, they need to be urgently transported to the hospital, where they will be observed and provided with medical help if health condition is at risk. At the same time, the amount of elderly people choosing to maintain their independent lifestyles is growing rapidly, which makes it harder for medical profes-sionals to follow changes and trends in patient’s health conditions outside hospital environment. However, remote monitoring can help to prevent described scenario, significantly reduce healthcare costs and at the same time maintain patient’s independent lifestyle [3]. Therefore, there is a clear demand in reliable multi-functional remote monitoring systems for elderly people, which collect and combine different sources of med-ical data corresponding to everyday routine of the monitored patient. In many cases, different components comprising the systems are dis-integrated and operating separately from each other. However, if we combine monitoring components (e.g. sensors, actuators) into smart en-vironments, we will be able to carry out observations for people with multiple chronic conditions at home. It will help to improve elderly pa-tient’s level of freedom and safety, which is one of the main issues in healthcare industry.

4 Chapter 1. Introduction

At teh same time fall incidents are considered to be one of the most common and dangerous risks among elderly population, with nearly half of nursing home residents and 30% of independently living people falling each year. Therefore, modern healthcare systems tend to integrate re-liable fall detection functionality into general monitoring framework. With the recent development on ICT market wearable sensors are of-ten deployed in conjunction with environmental devices to improve fall detection rates and minimize false alarms. In this case a multi-modal system requires a special fusion algorithm to combine all the active com-ponents.

Another source of motivation is a lack of contextual data in a vast majority of modern healthcare systems. Therefore, smart homes, with a capability to unobtrusively collect contextual data (e.g. radio-frequency identification (RFID) tags, pressure mats, switcher sensors and etc.) are essential sources of information. These data can be processed afterwards and infer real activities (e.g. cooking, sleeping, exercising), giving an ex-tra insight on physiological processes happening with elderly. However, there are no obvious solutions for integrating medical sensors into a smart home environment, which makes this area open for further research in-vestigations.

1.2

Problem description

As a response to the aging population, modern healthcare market pro-vides a wide range of medical devices for remote measuring of vital health parameters. Most of the equipment is programmed and exploit for spot checking and is not able to give a continuous overview of the patient’s health conditions. Moreover, different parameters are measured sepa-rately and monitoring process is not synchronized. At the same time, modern smartphones are equipped with advanced sensor functionality, which has a great potential for the healthcare, but is mostly exploited in game industry.

Assuming mentioned circumstances and previously conducted study focused on android based monitoring system for patients with chronic obtrusive disease [4] we can formulate two main challenges of presented research: (1) a lack of knowledge and sufficient expertize for continuous medical data analyses and (2) inefficient insight on the correlation be-tween measured parameters.

1.3 Research Hypothesis 5

Assuming formulated problems, current research investigates the pos-sibilities of continuous remote monitoring of elderly people in their home environment, collecting vital parameters and subsequent analyses to-wards correlation between acquired measurements. A special focus is made on multiple sources of medical data, different communication pro-tocols and variety of processing algorithms for alarm generation.

1.3

Research Hypothesis

We believe the future development of wireless monitoring is based on integration of unrelated data sources into a multi-modal system and switching between the components depending on particular situation. The result can be achieved by gradual development of independent com-ponents with a subsequent integration into a common framework with a generic processing algorithm. We believe these types of system are able to replace modern spot-checking health sensors, can be implemented with a low power consumption rate, improve reliability and increase the overall acceptance of health monitoring systems.

1.4

Research Questions

During development and investigation process we expect to answer a number of research questions to check the formulated hypothesis. Each question corresponds to particular challenges and demands in terms of remote monitoring domain and healthcare industry in general.

Question A

Which combination of sensors, actuators and measured parameters have the potential to provide sufficient amount of data for on-line and posterior analyses. We are looking for research solutions which will help to establish unobtrusive monitoring and at the same time measure vital parameters both outdoors and in home environment. Question B

Followong elaborated data collection process, we want to investi-gate what types of algorithms should be involved to perform anal-yses of the collected measurements.

4 Chapter 1. Introduction

At teh same time fall incidents are considered to be one of the most common and dangerous risks among elderly population, with nearly half of nursing home residents and 30% of independently living people falling each year. Therefore, modern healthcare systems tend to integrate re-liable fall detection functionality into general monitoring framework. With the recent development on ICT market wearable sensors are of-ten deployed in conjunction with environmental devices to improve fall detection rates and minimize false alarms. In this case a multi-modal system requires a special fusion algorithm to combine all the active com-ponents.

Another source of motivation is a lack of contextual data in a vast majority of modern healthcare systems. Therefore, smart homes, with a capability to unobtrusively collect contextual data (e.g. radio-frequency identification (RFID) tags, pressure mats, switcher sensors and etc.) are essential sources of information. These data can be processed afterwards and infer real activities (e.g. cooking, sleeping, exercising), giving an ex-tra insight on physiological processes happening with elderly. However, there are no obvious solutions for integrating medical sensors into a smart home environment, which makes this area open for further research in-vestigations.

1.2

Problem description

As a response to the aging population, modern healthcare market pro-vides a wide range of medical devices for remote measuring of vital health parameters. Most of the equipment is programmed and exploit for spot checking and is not able to give a continuous overview of the patient’s health conditions. Moreover, different parameters are measured sepa-rately and monitoring process is not synchronized. At the same time, modern smartphones are equipped with advanced sensor functionality, which has a great potential for the healthcare, but is mostly exploited in game industry.

Assuming mentioned circumstances and previously conducted study focused on android based monitoring system for patients with chronic obtrusive disease [4] we can formulate two main challenges of presented research: (1) a lack of knowledge and sufficient expertize for continuous medical data analyses and (2) inefficient insight on the correlation be-tween measured parameters.

1.3 Research Hypothesis 5

Assuming formulated problems, current research investigates the pos-sibilities of continuous remote monitoring of elderly people in their home environment, collecting vital parameters and subsequent analyses to-wards correlation between acquired measurements. A special focus is made on multiple sources of medical data, different communication pro-tocols and variety of processing algorithms for alarm generation.

1.3

Research Hypothesis

We believe the future development of wireless monitoring is based on integration of unrelated data sources into a multi-modal system and switching between the components depending on particular situation. The result can be achieved by gradual development of independent com-ponents with a subsequent integration into a common framework with a generic processing algorithm. We believe these types of system are able to replace modern spot-checking health sensors, can be implemented with a low power consumption rate, improve reliability and increase the overall acceptance of health monitoring systems.

1.4

Research Questions

During development and investigation process we expect to answer a number of research questions to check the formulated hypothesis. Each question corresponds to particular challenges and demands in terms of remote monitoring domain and healthcare industry in general.

Question A

Which combination of sensors, actuators and measured parameters have the potential to provide sufficient amount of data for on-line and posterior analyses. We are looking for research solutions which will help to establish unobtrusive monitoring and at the same time measure vital parameters both outdoors and in home environment. Question B

Followong elaborated data collection process, we want to investi-gate what types of algorithms should be involved to perform anal-yses of the collected measurements.

6 Chapter 1. Introduction

1. First part of the question will concern the ”on-line” processing performed during the monitoring stage

2. Alternatively we will investigate possible algorithms for pos-terior analyses applied to previously collected data

Question C

Another important research issue concerns the list of optimal out-puts determined to provide competence assistance to medical staff or notify patients directly. In this case, we want to investigate various user cases and monitoring scenarios which require different approaches in information delivery. Possible options can include: instant alarm indication, visualization feature or diagnose assis-tance.

Question D

Additionally we conduct a review study to give a better insight on multi-fusion based monitoring and fall detection solutions. A large number of healthcare device are already available for long-term monitoring and can be potentially integrated into a common network. It is not clear however which integration mechanism or fusion tecnique is the most applicable for this approach.

1.5

Thesis Outline

The rest of the thesis is organized as following:

Part I consists of five chapters. Chapter 1 provides a motivation for ini-tiated studies, gives a brief problem description, formulate main hypoth-esis and research questions. Chapter 2 presents background knowledge on research topic and describes the latest publications within wireless healthcare systems with a special focus on fall detection and smart home environments. In Chapter 3 we provide details on research methodology deployed to achieve goals and answer announced questions. Chapter 4 contains main thesis contributions listed according to their relevance to the published papers. Finally we draw conclusions and discuss possible future work in Chapter 5.

Part II presents technical contributions of the thesis in the form of three papers that are organized in Chapters 6 to 8.

Chapter 2

Wireless Monitoring

Particular circumstances including aging of population in developed coun-tries, increasing costs of primary healthcare and strong demand in inde-pendent living, have already evoked an intensive research work in remote monitoring area. Normally, these types of systems are subdivided into three major sections: sensor layer, communication layer and caregiver. We are particularly interested in the first two categories responsible for collecting, transferring and processing of the streaming data. Usually monitoring process involves both wearable and environmental sensors, collecting data for further processing and visualizing. However, these two types of information channels are managed separately and rarely presented as a combined structure.

2.1

Wearable Sensors

Latest wearable medical devices often operate in conjunction with smart-phones, which are playing a major role in the modern healthcare systems [3]. The list of possible applications has been growing along with market development: early detection of Alzheimer’s disease [5], face-to-face com-munication between doctor and patient [6], complex activity recognition [7] and medicine in-take assistance [8]. Modern smartphones, operating as wearable sensor can also be deployed as a communication entity for other medical devices providing a link between different types of data sources [9]. This particular idea was implemented and described in an ar-ticle by Kohei Arai [10], where blood pressure, body temperature, pulse

6 Chapter 1. Introduction

1. First part of the question will concern the ”on-line” processing performed during the monitoring stage

2. Alternatively we will investigate possible algorithms for pos-terior analyses applied to previously collected data

Question C

Another important research issue concerns the list of optimal out-puts determined to provide competence assistance to medical staff or notify patients directly. In this case, we want to investigate various user cases and monitoring scenarios which require different approaches in information delivery. Possible options can include: instant alarm indication, visualization feature or diagnose assis-tance.

Question D

Additionally we conduct a review study to give a better insight on multi-fusion based monitoring and fall detection solutions. A large number of healthcare device are already available for long-term monitoring and can be potentially integrated into a common network. It is not clear however which integration mechanism or fusion tecnique is the most applicable for this approach.

1.5

Thesis Outline

The rest of the thesis is organized as following:

Part I consists of five chapters. Chapter 1 provides a motivation for ini-tiated studies, gives a brief problem description, formulate main hypoth-esis and research questions. Chapter 2 presents background knowledge on research topic and describes the latest publications within wireless healthcare systems with a special focus on fall detection and smart home environments. In Chapter 3 we provide details on research methodology deployed to achieve goals and answer announced questions. Chapter 4 contains main thesis contributions listed according to their relevance to the published papers. Finally we draw conclusions and discuss possible future work in Chapter 5.

Part II presents technical contributions of the thesis in the form of three papers that are organized in Chapters 6 to 8.

Chapter 2

Wireless Monitoring

Particular circumstances including aging of population in developed coun-tries, increasing costs of primary healthcare and strong demand in inde-pendent living, have already evoked an intensive research work in remote monitoring area. Normally, these types of systems are subdivided into three major sections: sensor layer, communication layer and caregiver. We are particularly interested in the first two categories responsible for collecting, transferring and processing of the streaming data. Usually monitoring process involves both wearable and environmental sensors, collecting data for further processing and visualizing. However, these two types of information channels are managed separately and rarely presented as a combined structure.

2.1

Wearable Sensors

Latest wearable medical devices often operate in conjunction with smart-phones, which are playing a major role in the modern healthcare systems [3]. The list of possible applications has been growing along with market development: early detection of Alzheimer’s disease [5], face-to-face com-munication between doctor and patient [6], complex activity recognition [7] and medicine in-take assistance [8]. Modern smartphones, operating as wearable sensor can also be deployed as a communication entity for other medical devices providing a link between different types of data sources [9]. This particular idea was implemented and described in an ar-ticle by Kohei Arai [10], where blood pressure, body temperature, pulse

8 Chapter 2. Wireless Monitoring

rate EEG, calorie consumption and other sensors are attached to the human body. Measured data are transferred to mobile devices through Bluetooth and further to the Information Collection Center with the help of WiFi or Wireless LAN. Alternatively, we can establish connection to a smartphone device via ZigBee, which is a standard communication protocol for low-cost, low-power, wireless sensor and control networks. Matthias Wagner et al [11] developed a two-approach telemedical sys-tem focused on the measurement and evaluation of vital parameters, e.g. ECG, heart rate, heart rate variability, pulse oximetry, plethysmography and fall detection. According to the first approach, all the obtained pa-rameters are transferred to coordinator node via ZigBee technique, which as Bluetooth operates on the GHz radio frequency, but has a maximum data rate of 0.25 Mbps. Both protocols have been additionally improved to be used by medical devices via universal communication standards including ”Health Care” for ZigBee and Health Device Profile (HDP) for Bluetooth.

Apart from their communication functionality, mobile devices can serve as preprocessing tool during the monitoring. This feature allows to avoid overloading potential caregiver with unnecessary information or even trigger an early alarm in case of emergency [12]. Kozlovsky et al [13] developed an Android based mobile data acquisition (DAQ) so-lution, which collects personalized health information of the end-user, store, analyze and visualize it on the smart device and optionally sends it to the data center for further processing. The software even enables correlation analysis between the various sensor data sets. This option, however, is not feasible for all the mobile devices due to some differences in processing time and memory consumption.

In our research study we tend to investigate potential benefits of fus-ing unrelated sorts of data into a multi-modal framework. Therefore, in addition to medical entities based on smartphones, we deploy environ-mental sensors and involve them into a monitoring process. We select emergency situations associated with falls as a main application for de-veloping system, which can be potentially expanded to a larger scale of healthcare problems.

2.2 Fall Detection 9

2.2

Fall Detection

As it was previously discussed in Chapter 1, falls are among major prob-lems in modern healthcare and a serious threat for elderly population. As a result, most of the wireless monitoring systems tend to include automatic fall detection into their functionality. Modern smartphones are often equipped with a set of powerful sensor technology and start to play a significant role in healthcare development. Recent studies proved that accelerometer, gyroscope and magnetometer can comprise an inde-pendent fall detecting tool or be a part of the fall detection framework. Commonly, acceleration data is collected and stored on the smartphone with subsequent on-line or off-line processing depending on the current circumstances. Alternatively, some of the studies propose algorithms where contextual or visual data collected by environmental sensors is deployed to detect a fall. In this case obtrusiveness of the process is relatively low since patients do not require to wear any devices. At the same time, these type of systems are often facing privacy issues and re-quire additional ethical approve. Due to these reasons and complexity of the fall process in general several attempts were made to combine both types of data to improve overall performance of fall detection systems. In the following section we provide main fall characteristics, describe popular approaches and explain how fall detection can be included in a general monitoring model implemented in a smart home environemt.

2.2.1

Fall Characterisitcs

A fall is commonly defined as ”unintentionally coming to rest on the ground, floor, or other lower level”. Losing the balance and subsequent falling with the help of an assistant also considered as a fall [14]. Based on possible scenarios 4 main types of falls can be distinguished: (1) fall from sleeping, (2) fall from sitting, (3) fall from walking/standing and (4) fall from standing on support tools such as ladder. Each type has its’ own unique characteristics, which can help developers to adapt fall detector platforms to a wider spectrum of user requirements.

Typically all the modern fall detection systems can be split into 3 main classes depending on the sensor technology deployed for monitor-ing: wearable sensors, ambient sensors and vision-based sensors (see Fig-ure 2.1). Most of the wearable fall detectors are based on accelerometer

8 Chapter 2. Wireless Monitoring

rate EEG, calorie consumption and other sensors are attached to the human body. Measured data are transferred to mobile devices through Bluetooth and further to the Information Collection Center with the help of WiFi or Wireless LAN. Alternatively, we can establish connection to a smartphone device via ZigBee, which is a standard communication protocol for low-cost, low-power, wireless sensor and control networks. Matthias Wagner et al [11] developed a two-approach telemedical sys-tem focused on the measurement and evaluation of vital parameters, e.g. ECG, heart rate, heart rate variability, pulse oximetry, plethysmography and fall detection. According to the first approach, all the obtained pa-rameters are transferred to coordinator node via ZigBee technique, which as Bluetooth operates on the GHz radio frequency, but has a maximum data rate of 0.25 Mbps. Both protocols have been additionally improved to be used by medical devices via universal communication standards including ”Health Care” for ZigBee and Health Device Profile (HDP) for Bluetooth.

Apart from their communication functionality, mobile devices can serve as preprocessing tool during the monitoring. This feature allows to avoid overloading potential caregiver with unnecessary information or even trigger an early alarm in case of emergency [12]. Kozlovsky et al [13] developed an Android based mobile data acquisition (DAQ) so-lution, which collects personalized health information of the end-user, store, analyze and visualize it on the smart device and optionally sends it to the data center for further processing. The software even enables correlation analysis between the various sensor data sets. This option, however, is not feasible for all the mobile devices due to some differences in processing time and memory consumption.

In our research study we tend to investigate potential benefits of fus-ing unrelated sorts of data into a multi-modal framework. Therefore, in addition to medical entities based on smartphones, we deploy environ-mental sensors and involve them into a monitoring process. We select emergency situations associated with falls as a main application for de-veloping system, which can be potentially expanded to a larger scale of healthcare problems.

2.2 Fall Detection 9

2.2

Fall Detection

As it was previously discussed in Chapter 1, falls are among major prob-lems in modern healthcare and a serious threat for elderly population. As a result, most of the wireless monitoring systems tend to include automatic fall detection into their functionality. Modern smartphones are often equipped with a set of powerful sensor technology and start to play a significant role in healthcare development. Recent studies proved that accelerometer, gyroscope and magnetometer can comprise an inde-pendent fall detecting tool or be a part of the fall detection framework. Commonly, acceleration data is collected and stored on the smartphone with subsequent on-line or off-line processing depending on the current circumstances. Alternatively, some of the studies propose algorithms where contextual or visual data collected by environmental sensors is deployed to detect a fall. In this case obtrusiveness of the process is relatively low since patients do not require to wear any devices. At the same time, these type of systems are often facing privacy issues and re-quire additional ethical approve. Due to these reasons and complexity of the fall process in general several attempts were made to combine both types of data to improve overall performance of fall detection systems. In the following section we provide main fall characteristics, describe popular approaches and explain how fall detection can be included in a general monitoring model implemented in a smart home environemt.

2.2.1

Fall Characterisitcs

A fall is commonly defined as ”unintentionally coming to rest on the ground, floor, or other lower level”. Losing the balance and subsequent falling with the help of an assistant also considered as a fall [14]. Based on possible scenarios 4 main types of falls can be distinguished: (1) fall from sleeping, (2) fall from sitting, (3) fall from walking/standing and (4) fall from standing on support tools such as ladder. Each type has its’ own unique characteristics, which can help developers to adapt fall detector platforms to a wider spectrum of user requirements.

Typically all the modern fall detection systems can be split into 3 main classes depending on the sensor technology deployed for monitor-ing: wearable sensors, ambient sensors and vision-based sensors (see Fig-ure 2.1). Most of the wearable fall detectors are based on accelerometer

10 Chapter 2. Wireless Monitoring

data and operating with posture and motion of the patients body.

Vision-based sensors

Wearable Sensors Ambient sensors

Posture Motion

Body shape change

3-D head change

Inactivity Posture Presence

Fall Detection

Figure 2.1: Fall detection classification

They can additionally be subdivided into thresholding or machine learning methods according to the processing algorithm they deploy. Ac-celeration data collected during the fall in different directions is demon-strated on Figure 2.2. Each line represents raw, pitch or yaw of the smartphones coordinate axis, has its unique variation and can distinc-tively depict three different fazes of every fall motion: (1) pre-fall, (2) impact, (3) after-fall phase.

50 100 150 200 250 300 −10 0 10 Time, hours Average activity 50 100 150 200 250 300 350 400 −10 0 10 Time, hours Average activity 50 100 150 200 250 300 −10 0 10 Measurements Average activity

Figure 2.2: Smartphone coordinate axis

Alternative fall detection methods are based on contextual data and

2.2 Fall Detection 11

deploy modern vision or ambient techniques to detect a fall. In this case collected measurements (video stream, sound etc) are transfered to a remote device and inspected for possible emergency situations associ-ated with falls. In the vast majority of the context-aware or vision-based systems falls are detected off-line with the help of statistical or machine learning algorithms.

Each of the presented approaches still gives a significant amount of false positive alarms while operating independently. It is therefore im-portant to integrate additional sensor functionality in order to improve reliability of fall detection systems. This trend is becoming popular and addressed as multi-sensor fusion based fall detection. In this case, sev-eral sensor channels are deployed to collect data which is later fused on a processing level. In the following sections we continue to describe each type of fall detection approach in particular and discuss possible solutions for a multimodal framework fusing both techniques.

2.2.2

Context-Aware Fall Detection

Contextual data has been recently deployed to perform fall detection and activity recognition of elderly people in their home environment. Brulin et al.[15] described an approach, with the main idea to fuse different types of data source channels into a special architecture, acquiring infor-mation from RIP detectors, thermopile or cameras. This system can also perform on-line or posterior processing to derive posture or orientation of the user and trigger an alarm in case of fall risk. Various attempts were made to improve this process by introducing additional sources of data like surrounding audio captured by microphone arrays [16, 17] or cur-rent location of the user [18]. Alternatively, RGBD-camera is deployed in study by BinBing Ni et al [19] for hospital fall prevention. Another recent study demonstrates how fall prevention system makes use of col-lected data from sensors in order to control and advice patients or even to give instructions to treat an abnormal condition and reduce the fall risk [20]. In this case monitoring and processing data from sensors is performed by a smartphone that will issue warnings to the user and in emergency situations send them to a caregiver. Moreover, relationship between acceleration of body’s center of gravity during sit-to-walk mo-tion and a process of falling is investigated by Shiozawa, N et al. [21]. The result of discriminant analysis by using indexes with a significant difference revealed a 90.3% correct prediction rate for falling. However,

10 Chapter 2. Wireless Monitoring

data and operating with posture and motion of the patients body.

Vision-based sensors

Wearable Sensors Ambient sensors

Posture Motion

Body shape change

3-D head change

Inactivity Posture Presence

Fall Detection

Figure 2.1: Fall detection classification

They can additionally be subdivided into thresholding or machine learning methods according to the processing algorithm they deploy. Ac-celeration data collected during the fall in different directions is demon-strated on Figure 2.2. Each line represents raw, pitch or yaw of the smartphones coordinate axis, has its unique variation and can distinc-tively depict three different fazes of every fall motion: (1) pre-fall, (2) impact, (3) after-fall phase.

50 100 150 200 250 300 −10 0 10 Time, hours Average activity 50 100 150 200 250 300 350 400 −10 0 10 Time, hours Average activity 50 100 150 200 250 300 −10 0 10 Measurements Average activity

Figure 2.2: Smartphone coordinate axis

Alternative fall detection methods are based on contextual data and

2.2 Fall Detection 11

deploy modern vision or ambient techniques to detect a fall. In this case collected measurements (video stream, sound etc) are transfered to a remote device and inspected for possible emergency situations associ-ated with falls. In the vast majority of the context-aware or vision-based systems falls are detected off-line with the help of statistical or machine learning algorithms.

Each of the presented approaches still gives a significant amount of false positive alarms while operating independently. It is therefore im-portant to integrate additional sensor functionality in order to improve reliability of fall detection systems. This trend is becoming popular and addressed as multi-sensor fusion based fall detection. In this case, sev-eral sensor channels are deployed to collect data which is later fused on a processing level. In the following sections we continue to describe each type of fall detection approach in particular and discuss possible solutions for a multimodal framework fusing both techniques.

2.2.2

Context-Aware Fall Detection

Contextual data has been recently deployed to perform fall detection and activity recognition of elderly people in their home environment. Brulin et al.[15] described an approach, with the main idea to fuse different types of data source channels into a special architecture, acquiring infor-mation from RIP detectors, thermopile or cameras. This system can also perform on-line or posterior processing to derive posture or orientation of the user and trigger an alarm in case of fall risk. Various attempts were made to improve this process by introducing additional sources of data like surrounding audio captured by microphone arrays [16, 17] or cur-rent location of the user [18]. Alternatively, RGBD-camera is deployed in study by BinBing Ni et al [19] for hospital fall prevention. Another recent study demonstrates how fall prevention system makes use of col-lected data from sensors in order to control and advice patients or even to give instructions to treat an abnormal condition and reduce the fall risk [20]. In this case monitoring and processing data from sensors is performed by a smartphone that will issue warnings to the user and in emergency situations send them to a caregiver. Moreover, relationship between acceleration of body’s center of gravity during sit-to-walk mo-tion and a process of falling is investigated by Shiozawa, N et al. [21]. The result of discriminant analysis by using indexes with a significant difference revealed a 90.3% correct prediction rate for falling. However,

12 Chapter 2. Wireless Monitoring

there is still a relatively high level of false alarm generated associated with the context-based detection and prevention systems, which can be potentially improved by integrating with a wearable device.

It was previously discussed that wearable sensors with inbuilt ac-celerometer can serve as an effective healthcare device. Most of them have recently been deployed for accurate fall detection [6] demonstrat-ing dignified results durdemonstrat-ing evaluation process. We are interested in accelerometer-based sensor both as an individual component and as a part of the multi-modal system, where it is combined with environmen-tal sensors. In our case a wearable device is replaced with a smartphone, which can serve as fall detector sensor and gateway mode at the same time.

2.2.3

Smartphone-based Fall Detection

With the recent development on mobile market, smartphones start to play an important role in modern healthcare systems [3]. Latest ver-sions equipped with an accelerometer sensor are commonly used as fall detection tools [22, 23, 24, 25]. In this case they replace both process-ing mode and a communication tool while maintainprocess-ing relatively small size. A choice of processing algorithm depends on final application of the system and varies in different studies. Some of the recent imple-mentation methods apply Gaussian distribution of clustered knowledge [26], neural network [27] and machine learning techniques [28]. However, most of them are initially based on three essential parameters associated with falls: impact, velocity and posture. According to the recent article, combining impact and posture while analyzing the fall case is enough to create a reliable algorithm [29].

Based on research questions formulated earlier, we tend to develop a fall detection system and investigate its further integration into a full-scale monitoring system with additional sensor functionality includ-ing medical devices and environmental sensors. Similar approach was adopted in several studies with intention to combine contextual data with essential accelerometer measurements exploiting inertia and loca-tion sensors [30]. Qiang Li et al in [31] investigate a novel fall detec-tion method that utilizes acceleradetec-tion, posture and context informadetec-tion, where context can be presented by environmental sensors (room loca-tion or furniture posiloca-tions) and personal profiles (e.g. health status and age). Wireless accelerometer, 3-D camera and microphone are being

si-2.3 Smart Home Environment 13

multaneously processed by Leone et al to reach a better result in fall risk assessment [32]. All the presented studies, however, are lacking a reliable fusing technique to combine processing results from independent components. In work by Zhang et al [33] an off-the-shelf programmable sensing platform called Sun SPOT is used for data recording. Context information is presented in several categories, covering main aspects of elderly living: (1) physical activity; (2) physiological condition; (3) per-sonal health record; and (4) location. Each category represents a sep-arate variable or processing result (including fall alarm from isolated algorithm) and merged into a Bayesian network for further statistical analyses. In the current research, we make an attempt to develop an in-dependently operating fall detection system, which can be easily merged with other types of sensors and integrated into a larger monitoring en-vironment. Therefore in Paper B, we describe a method for managing multi-variant healthcare data with Dynamic Bayesian Network, and pro-pose the framework of the developed system.

2.3

Smart Home Environment

Reliable fall detection is one of the steps towards universal monitor-ing system for elderly people with different types of disease. It should be combined with other sources of medical information channels in or-der to give a better insight on patients medical conditions. This can be successfully implemented via modern smart home environments as a multi-modal platform with fusing capabilities. Smart homes in general are a part of the Ambient Assisted Living area, responsible for continu-ous monitoring of elderly people in comfortable home environment. In recent years, an increasing number of projects have been based on this approach implying various components and applications. Special nutri-tion advisor was proposed as an attempt to improve physical condinutri-tion for elderly people with diabetes [34]. It demonstrates the possibility to make nutritional management much more effective through deploying Ambient Intelligence systems. Moreover, Juan A Botia et al. [35] pre-sented their Necesity system with adaptive monitoring capabilities and exhaustive evaluation methodology which was integrated in the develop-ment process. Jer-Vui Lee et al. [36] make an attempt to build a smart elderly home based on an android device, which is utilized as a 3-axial accelerometer device to detect a fall of the carrier.

12 Chapter 2. Wireless Monitoring

there is still a relatively high level of false alarm generated associated with the context-based detection and prevention systems, which can be potentially improved by integrating with a wearable device.

It was previously discussed that wearable sensors with inbuilt ac-celerometer can serve as an effective healthcare device. Most of them have recently been deployed for accurate fall detection [6] demonstrat-ing dignified results durdemonstrat-ing evaluation process. We are interested in accelerometer-based sensor both as an individual component and as a part of the multi-modal system, where it is combined with environmen-tal sensors. In our case a wearable device is replaced with a smartphone, which can serve as fall detector sensor and gateway mode at the same time.

2.2.3

Smartphone-based Fall Detection

With the recent development on mobile market, smartphones start to play an important role in modern healthcare systems [3]. Latest ver-sions equipped with an accelerometer sensor are commonly used as fall detection tools [22, 23, 24, 25]. In this case they replace both process-ing mode and a communication tool while maintainprocess-ing relatively small size. A choice of processing algorithm depends on final application of the system and varies in different studies. Some of the recent imple-mentation methods apply Gaussian distribution of clustered knowledge [26], neural network [27] and machine learning techniques [28]. However, most of them are initially based on three essential parameters associated with falls: impact, velocity and posture. According to the recent article, combining impact and posture while analyzing the fall case is enough to create a reliable algorithm [29].

Based on research questions formulated earlier, we tend to develop a fall detection system and investigate its further integration into a full-scale monitoring system with additional sensor functionality includ-ing medical devices and environmental sensors. Similar approach was adopted in several studies with intention to combine contextual data with essential accelerometer measurements exploiting inertia and loca-tion sensors [30]. Qiang Li et al in [31] investigate a novel fall detec-tion method that utilizes acceleradetec-tion, posture and context informadetec-tion, where context can be presented by environmental sensors (room loca-tion or furniture posiloca-tions) and personal profiles (e.g. health status and age). Wireless accelerometer, 3-D camera and microphone are being

si-2.3 Smart Home Environment 13

multaneously processed by Leone et al to reach a better result in fall risk assessment [32]. All the presented studies, however, are lacking a reliable fusing technique to combine processing results from independent components. In work by Zhang et al [33] an off-the-shelf programmable sensing platform called Sun SPOT is used for data recording. Context information is presented in several categories, covering main aspects of elderly living: (1) physical activity; (2) physiological condition; (3) per-sonal health record; and (4) location. Each category represents a sep-arate variable or processing result (including fall alarm from isolated algorithm) and merged into a Bayesian network for further statistical analyses. In the current research, we make an attempt to develop an in-dependently operating fall detection system, which can be easily merged with other types of sensors and integrated into a larger monitoring en-vironment. Therefore in Paper B, we describe a method for managing multi-variant healthcare data with Dynamic Bayesian Network, and pro-pose the framework of the developed system.

2.3

Smart Home Environment

Reliable fall detection is one of the steps towards universal monitor-ing system for elderly people with different types of disease. It should be combined with other sources of medical information channels in or-der to give a better insight on patients medical conditions. This can be successfully implemented via modern smart home environments as a multi-modal platform with fusing capabilities. Smart homes in general are a part of the Ambient Assisted Living area, responsible for continu-ous monitoring of elderly people in comfortable home environment. In recent years, an increasing number of projects have been based on this approach implying various components and applications. Special nutri-tion advisor was proposed as an attempt to improve physical condinutri-tion for elderly people with diabetes [34]. It demonstrates the possibility to make nutritional management much more effective through deploying Ambient Intelligence systems. Moreover, Juan A Botia et al. [35] pre-sented their Necesity system with adaptive monitoring capabilities and exhaustive evaluation methodology which was integrated in the develop-ment process. Jer-Vui Lee et al. [36] make an attempt to build a smart elderly home based on an android device, which is utilized as a 3-axial accelerometer device to detect a fall of the carrier.

14 Chapter 2. Wireless Monitoring

We tend to utilize smart home environment in order to build a long-term monitoring platform for elderly people. At the same time, some of its capabilities can be used to collect contextual data, which makes it an additional source of information. Villarrubia et al. [37] make an attempt to incorporate image processing and artificial techniques based on PANGEA (Platform for the Automatic Construction of Organiztions of Intelligent Agents) plaform. The system is presented in a case study designed using different agents and sensors responsible for providing user support at home in the event of incidents or emergencies. Developers of the eCAALYX project [3] (Enhanced Complete Ambient Assisted Liv-ing Experiment) take 24/7 monitorLiv-ing of healthy older people one step further by refining it and making it available to older people with multi-ple chronic disease. A particular effort is made on communication with the user deploying various sorts of interactive devices: TV-based Set Top Box system, Customer-Premises Equipment and interactive TV. The key idea is to extend independent life at home and avoid hospitalization for longer periods.

The number of publications in healthcare domain is expected to grow rapidly. There are both successful implementations and studies which require additional research effort. The obvious trend observed in most of the recent publications can be characterized by combining unrelated source of data into integrated framework. However, it is still not clear how to process collected measurements or which set up will be most effective in the biggest amount of monitoring instances. Therefore, we believe it is important to (1) combine both wearable and contextual data to be able to adjust developed system for different types of users and monitoring scenarios, (2) create a flexible integration platform to be able to add/remove sensors depending on particular user requirements. In this case an efficient and flexible algorithm is required for fusing var-ious sources of data before the processing stage. Therefore, as a part of the research process, we conducted a literature search resulted in a review paper on recent studies within the multi-sensor fusion based fall detection. Combination of contextual and wearable data based on smart home platform for effective detection of emergency situations associated with falls is a step towards the full-scale monitoring system for elderly.

Chapter 3

Methodology

3.1

Research Approach

As it was previously formulated in Section 1.2, main challenges of the proposed study include (1) a lack of knowledge and sufficient expertize for continuous medical data analyses and (2) a lack of efficient fusing algorithm for managing unrelated data sources (3) inefficient insight on the correlation between various parameters. All challenges represent dif-ferent types of analyses interacting between each other, and therefore can be approached with cross-disciplinary research. A lack of knowledge for continuous data analyses can be overcome by introducing additional data sources to the monitoring process. In order to provide a better insight on a correlation between measured parameters we can deploy multi-sensor fusion algorithm and combine different types of data into a single source channel. The choice of algorithm in this case can be based on literature search and review of the recent studies within multi-modal monitoring area. Therefore, we commenced a multi-disciplinary research work in-cluding components like sensor management, signal processing, android development, artificial intelligence and medical knowledge expertise.

Each area corresponds to a particular stage of the monitoring pro-cess and therefore requires independent research approach, hardware and software configurations. The vital component of any remote diagnostic system is a sensor layer, which is responsible for the continuous data collection. Rapidly growing market of medical devices allow us to de-ploy different types of sensors and expand the number of vitals sings to

14 Chapter 2. Wireless Monitoring

We tend to utilize smart home environment in order to build a long-term monitoring platform for elderly people. At the same time, some of its capabilities can be used to collect contextual data, which makes it an additional source of information. Villarrubia et al. [37] make an attempt to incorporate image processing and artificial techniques based on PANGEA (Platform for the Automatic Construction of Organiztions of Intelligent Agents) plaform. The system is presented in a case study designed using different agents and sensors responsible for providing user support at home in the event of incidents or emergencies. Developers of the eCAALYX project [3] (Enhanced Complete Ambient Assisted Liv-ing Experiment) take 24/7 monitorLiv-ing of healthy older people one step further by refining it and making it available to older people with multi-ple chronic disease. A particular effort is made on communication with the user deploying various sorts of interactive devices: TV-based Set Top Box system, Customer-Premises Equipment and interactive TV. The key idea is to extend independent life at home and avoid hospitalization for longer periods.

The number of publications in healthcare domain is expected to grow rapidly. There are both successful implementations and studies which require additional research effort. The obvious trend observed in most of the recent publications can be characterized by combining unrelated source of data into integrated framework. However, it is still not clear how to process collected measurements or which set up will be most effective in the biggest amount of monitoring instances. Therefore, we believe it is important to (1) combine both wearable and contextual data to be able to adjust developed system for different types of users and monitoring scenarios, (2) create a flexible integration platform to be able to add/remove sensors depending on particular user requirements. In this case an efficient and flexible algorithm is required for fusing var-ious sources of data before the processing stage. Therefore, as a part of the research process, we conducted a literature search resulted in a review paper on recent studies within the multi-sensor fusion based fall detection. Combination of contextual and wearable data based on smart home platform for effective detection of emergency situations associated with falls is a step towards the full-scale monitoring system for elderly.

Chapter 3

Methodology

3.1

Research Approach

As it was previously formulated in Section 1.2, main challenges of the proposed study include (1) a lack of knowledge and sufficient expertize for continuous medical data analyses and (2) a lack of efficient fusing algorithm for managing unrelated data sources (3) inefficient insight on the correlation between various parameters. All challenges represent dif-ferent types of analyses interacting between each other, and therefore can be approached with cross-disciplinary research. A lack of knowledge for continuous data analyses can be overcome by introducing additional data sources to the monitoring process. In order to provide a better insight on a correlation between measured parameters we can deploy multi-sensor fusion algorithm and combine different types of data into a single source channel. The choice of algorithm in this case can be based on literature search and review of the recent studies within multi-modal monitoring area. Therefore, we commenced a multi-disciplinary research work in-cluding components like sensor management, signal processing, android development, artificial intelligence and medical knowledge expertise.

Each area corresponds to a particular stage of the monitoring pro-cess and therefore requires independent research approach, hardware and software configurations. The vital component of any remote diagnostic system is a sensor layer, which is responsible for the continuous data collection. Rapidly growing market of medical devices allow us to de-ploy different types of sensors and expand the number of vitals sings to