Smart Aging: Enhancing Elderly well-Being Through App Based Activity Monitoring

PROJECT- REPORT
Agenda

 Introduction
 Problem statement
 Methodology/Approach
 Financial Analysis
 Result/Findings
 Future Scope
 Conclusion
 References
Introduction-
In the twenty-first century, geriatric care has attracted unprecedented attention
in almost all research fields to address the potential challenges related to
elderly people. The elderly population (aged 60 years or above) in India
accounted for 7.4% of the total population in 2001, 8.6% (104 million; 53
million females and 51 million males) in 2011, and has been projected to
increase to 19% by the year 2050. And thus leading to a growing need for
innovative solutions to support the health and well-being of elderly individuals.
Smart technology, including mobile applications, offers promising opportunities
to address these challenges. This project focuses on the use of app-based
activity monitoring to enhance the well-being of seniors, a concept referred to
as "Smart Aging."
As people aged, they often experience a decline in physical activity and an
increased risk of health issues such as cardiovascular disease, diabetes,
dementia and many more. Monitoring daily activities, including physical
activity levels, sleep patterns, and medication adherence, can provide valuable
insights into the health and well-being of elderly individuals. App-based
solutions offer a convenient and non-intrusive way to track these metrics,
enabling caregivers and healthcare professionals to intervene early if
necessary.
The primary objective of our project is to explore the potential of app-based
activity monitoring to improve the well-being of geriatric individuals. This
project is significant for several reasons. Firstly, it addresses an ongoing need to
support the health and well-being of the aging population. Secondly, it
leverages the potential of smart technology to provide personalized and
accessible healthcare solutions. Lastly, it contributes to the growing field of
Smart Aging, offering insights that can inform future research and policy
initiatives.
The report is organized into several sections, including a problem statement,
methodology, Result, Financial analysis, future scope and conclusions. Each
section provides detailed insights into the project's objectives, methods,
results, and implications for enhancing the well-being of elderly individuals
through app-based activity monitoring.
Problem Statement :
In our scenario, It has been assumed that a geriatric person is living alone in an
apartment. Moreover, the room plan within the apartment is known
beforehand. The problem in a single sentence is to build a low-cost solution for
identifying the daily activities of the person in real time. The first challenge is to
prepare a kit for sensing the presence of the concerned person in a specific
location within the flat. The necessary ambient sensors and the strategic
locations within the apartment need to be identified and the sensed data need
to be properly stored within a server for subsequent analysis. The second
challenge is to analyze the sensed data through machine learning and deep
learning techniques for activity detection or recognition.
Key considerations for our project include:
 Activity Monitoring: The app should be able to track various activities
such as BADL like Sleeping, Meal Preparation, Relax etc.
 Health Tracking: Integration with health monitoring devices (e.g., heart
rate monitors, blood pressure monitors) to track vital signs and health
indicators.
 Fall Detection: Incorporation of sensors or algorithms to detect falls and
alert caregivers or emergency services.
 Behavioral Analysis: Analyzing patterns in daily activities to detect
changes that may indicate health issues or decline in well-being.
 User Interface: A user-friendly interface that is accessible to elderly
individuals with different levels of technological proficiency.
So these are the things on which we have focussed in our project.
Methodology/ Approach:
We have categorised our project development work in stages and these are as
follows:-
 Sensor and other hardware identification
 Location Determination
 Data Acquisition and
 Data Analysis
Sensor And other hardware Identification : Identification of low-cost motion
sensors at the targeted area. We are more interested in using ambient
sensors (like motion sensor) that’s why we intend to take minimum support
from the wearable sensors. Based on our project requirement we have
specified sensors as well as other hardware parts and these are as follows:-

Product Usage Range /Capacity Total Count

Location Determination : It is one of the most important stage of our project

development work. We have to find some strategic locations so that the
identified sensors can be installed for data sensing. And that so our aim is to
place optimal number of sensors so that it covers the all strategic locations and
avoid the high cost of our project as well.
There are several factors which affects the sensor placement and these are as
follows:-
 Monitoring Goals: Identify the specific parameters we want to monitor,
such as physical activity, sleep patterns or basic activities. The location of
sensors will depend on these goals.
  User Comfort and Convenience: Sensors should be placed in a way that is
comfortable and convenient for the user.
 Accuracy and Reliability: Sensors should be placed in locations that
provide accurate and reliable data. For example, a motion sensor should
be placed in a position where it can accurately detect movement.
Data Acquisition: Now after sensors are placed we come on to the next step
[Link] collection and data processing.
When motion is sensed through any of the sensors, it sends all the data and
then data is stored in any repository. There are several options for storing data
but we are storing it in SQL server
Data Analysis: After we are done with data acquisition phase , we now start
data processing in real time. When we get several datapoints continuously, we
segment some datapoints and then recognition part is done in which any of the
machine learning algorithm or deep learning algorithm is used . Here In our
case we have used SVM machine learning algorithm.
And Finally, the outcome after data analysis is a daily activity chart of the
concerned person

SMART HOME

SENSOR STORAGE OF
DATA DATA
COLLECTION

DATA DATA DATA

RECOGNITION SEGMENTATION PROCESSING
FIG: Flowchart digram of how this model works

Financial Analysis:
For MQTT Server Hosting, we can host an MQTT broker on our own server
hardware, including a Raspberry Pi. Since, most brokers are free and open
source we can make it free or as low as possible. If we goes for cloud based
server, HiveMQ provides services with $0.34/hour + $0.80/million messages.

For Django Server Hosting, there are cost-effective server hosting providers as
low as $4/month. For example, Hostinger provides hosting for as low as INR
285/month.

Results/ Findings:

Future Scope:
This framework is expected to support multiple applications related to assisted
living solutions for the aging population in the upcoming times
Some of these include –
  It will be extended to multiple-people environments with minimal usage
of wearable sensors.
 The system will not only provide solutions to healthy persons, or mildly
cognitively impaired persons but also, it will help visually impaired
people and persons with special abilities.
 Also, some advance sensors will be placed for better data capture.

Conclusion:
The solution is designed to help community-dwelling older adults remain
safely in their homes and delay or prevent institutionalization. The solution
provides a range of services: starting from giving general assistance to
executing BADL (Basic Activities of Daily Living) and IADL (Instrumental
Activities of Daily Living) to take care of the physical and mental health of the
aged people. The solution enhances their confidence to stay independent for
longer. Also, it reduced the bed occupancy in the hospital. Also, geriatric people
gain more control over their social responsibility and local aspects of their
communities.

References:
 Kleinberger, T., Becker, M., Ras, E., Holzinger, A., M ̈ uller, P.: Ambient
intelligence in assisted living: enable elderly people to handle future
interfaces. In: International Conference on Universal Access in Human-
computer Interaction, pp. 103–112 (2007). Springer
 Thapliyal, H., Nath, R.K., Mohanty, S.P.: Smart home environment for
mild cognitive impairment population: Solutions to improve care and
quality of life. IEEE Consumer Electronics Magazine 7(1), 68–76 (2017).
 Arifoglu, D., Bouchachia, A.: Activity recognition and abnormal
behaviour detection with recurrent neural networks. Procedia Computer
Science 110, 86–93 (2017).
 Soto-Mendoza, V., Garc ́ıa-Mac ́ıas, J.A., Ch ́avez, E., Gomez-Montalvo,
J.R., Quintana, E.: Detecting abnormal behaviours of institutionalized
older adults through a hybrid-inference approach. Pervasive and Mobile
Computing
 Forkan, A.R.M., Khalil, I., Tari, Z., Foufou, S., Bouras, A.: A context- aware
approach for long-term behavioural change detection and abnormality
prediction in ambient assisted living. Pattern Recognition 48(3), 628–
641(2015)