Third Generation Teleassistance - Intelligent Monitoring Makes the Difference
-
Upload
eloisa-vargiu -
Category
Health & Medicine
-
view
174 -
download
0
Transcript of Third Generation Teleassistance - Intelligent Monitoring Makes the Difference
Third Generation TeleassistanceIntelligent Monitoring Makes the Difference
Xavier Rafael-Palou, Carme Zambrana, Stefan Dauwalder,
Enrique de la Vega, Eloisa Vargiu, and Felip Miralles
The Problem
Healthy Ageing
Elderly people aim to preserve their independence and autonomy at their own home as long as possible
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
By 2020, around 25% of the EU population will be over 65
People aged from 65-80 will rise by nearly 40% between 2010-2030
From 2012, the over-60 population will increase by about 2 million people a year
The median age of the EU population increased from 35.2 years in 1990, to 40.9 years by 2010
IT IS URGENT TO PROMOTE HELTHY AGEING THROUGH ICT SOLUTIONS
The Problem
Towards Third Generation Teleassistance
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
1st generation
• Reactive• No follow-up• Not integrated into
the HIS
The Problem
Towards Third Generation Teleassistance
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
1st generation
HELP!
• Reactive• No follow-up• Not integrated into
the HIS
2nd generation
• Reactive• Follow-up• Partially integrated
into the HIS
The Problem
Towards Third Generation Teleassistance
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
3rd generation
• Proactive• Totally automatic• Smart follow-up• Integrated into the
HISHEL!
AI Environment
Techniques, Methods, Tools
Machine learning
Data mining
Ambient intelligence
Monitoring
Knowledge and reasoning
Planning and Scheduling
Robotics
Computer vision
Personalization and profiling
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
Coordination
Self-organization
Self-adaptation
Collaborative filtering
Sensor networks
Multi-agent systems
Decision support systems
Recommender systems
AI Environment
Techniques, Methods, Tools
Machine learning
Data mining
Ambient intelligence
Monitoring
Knowledge and reasoning
Planning and Scheduling
Robotics
Computer vision
Personalization and profiling
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
Coordination
Self-organization
Self-adaptation
Collaborative filtering
Sensor networks
Multi-agent systems
Decision support systems
Recommender systems
The Proposed Solution
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
The Proposed Solution
Outline
Intelligent monitoring for…
Improving Sensor Reliability
Activity Recognition
Supporting carers
Assessing quality of life
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
eKauri: the teleassistance system
The Proposed Solution
Improving Sensor Reliability
High-level classifier SVM (RBF; γ = 1.0; C = 0.452)
Low-level classifier Novelty detection approach
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
The hierarchical approach
Rule-based Hierarchical Improvement
Accuracy 0.80 0.95 15%
Precision 0.68 0.94 26%
Recall 0.71 0.91 20%
F1 0.69 0.92 23%
The Proposed Solution
Sleeping Recognition
Supervised classifier: SVM (RBF; γ = 1.0; C = 1.0)
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
Sleeping recognition is aimed at reporting the time when the user went to sleep and woke up
the number of sleeping activityhours
the number of rest hours
The Proposed Solution
Supporting carers
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
The role of carers is essential for remotely assisting people that need assistance
The Proposed Solution
Supporting carers
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
The role of carers is essential for remotely assisting people that need assistance
The Proposed Solution
Assessing Quality of Life
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
We defined a Visual Analogic Scale QoL questionnaire composed of the following items MOOD
HEALTH
MOBILITY
SATISFACTION WITH CARE
USUAL ACTIVITIES (SLEEPING)
PAIN/DISCOMFORT
Item Accuracy
MOBILITY 0.76
SLEEPING 0.72
MOOD 0.81
Preliminary results
Conclusions
Third Generation Teleassistance
August 30, 2016 AI: Context – Multimodal Analytics – Internet of Things
ECAI 2016
Time is ripe to adopt IoT in the dependency care sector
Intelligent monitoring makes the difference in providing feedback to the users
User-centered solutions must be taken into account, involving final users in each phase of the development
We need concrete solutions that give a twist in adapting innovative strategies