See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/361276842

Smart Reminder SOS & Emergency Detection Device

Conference Paper · June 2022

DOI: 10.1109/ICDCECE53908.2022.9793171

CITATIONS READS

5 2,958

5 authors, including:

Madhumala R B Shashikala H K
Dayananda Sagar Institutions Jain University
29 PUBLICATIONS 111 CITATIONS 45 PUBLICATIONS 166 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Shashikala H K on 14 June 2022.

The user has requested enhancement of the downloaded file.

 2022 IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE)

Smart Reminder SOS & Emergency Detection

Device
2022 IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE) | 978-1-6654-8316-2/22/$31.00 ©2022 IEEE | DOI: 10.1109/ICDCECE53908.2022.9793171

Shashikala H.K Madhumala RB Chanchani keerthana

Department of Computer Science and Department of Computer Science and Department of Computer Science and
Engineering, Engineering, Engineering
JAIN (Deemed-to-be University), JAIN (Deemed-to-be University), JAIN (Deemed-to-be University)
Bangalore, India. Bangalore, India Bangalore, India
[email protected] [email protected] [email protected]

Priyanka.S Rangareddygari Meghana Yarragunta Thanmai

Department of Computer Science and Department of Computer Science and Department of Computer Science and
Engineering, Engineering, Engineering,
JAIN (Deemed-to-be University), JAIN (Deemed-to-be University), JAIN (Deemed-to-be University),
Bangalore, India. Bangalore, India. Banglore, India
[email protected] [email protected] [email protected]

Abstract—The project was inspired by three factors: the lack equipped with a little microprocessor for taking sensor
of a dedicated device as a reminder for chronic patients' feedings, running algorithms, and determining the nearest
medicine timings, the lag of an easily accessible SOS button, and hospital using a GPS tracker and informing local guardians
the third and most important, the lag of communication between through email [2]. The two other aspects of this device need
the time of the accident and contact with guardians or hospitals. the use of IOT in this project. The device must link to a
According to a study, approximately 85% of deaths caused by mailing, phoning, and texting system through the internet and
small and major heart attacks, strokes, and shortness of breath requires sensors and switches to transfer the data due to the
could be averted simply by treating them right away. As a result, availability of an auto-reminder system and an SOS button
a readily accessible SOS button would sufficient. Apart from
for emergency help. Rather than immediate analysis and
that, statistics show that not taking medications on time and
taking the wrong medicine at the wrong time are both harmful
detection, both study publications offer a way of prediction.
to one's health. In such a situation, a dedicated reminder device They employ data such as traffic data to detect a larger
that can remind of medicine names at specific times is required. number of autos [3].
Furthermore, according to News18, 36,579 people die each year They identify accident hotspots and pre-inform the driver
in vehicle accidents, 36,213 in two-wheeler accidents, 31,997 in about them in an unusual location or utilizing pre-saved data.
truck and lorry accidents, and almost 45,000 individuals in other What we've created is a full solution with no drawbacks such
means of transportation. Nearly 76 percent of deaths in each as data pre-saving, internet access requirements, or the need
vehicle are caused by a delay in getting the patient to the
for high speed. In addition, we have the following two
hospital. In half of the cases, guardians were unable to reach the
injured or the victim was not assisted in arriving at the hospital
features [4] Due to the fact that the majority of people suffer
on time. In such a situation, some gadget or technology that can from a sub-health problem or chronic disease, research into
detect any collision and alert the nearest hospital and local sensing and managing personal health is quickly expanding.
guardians as soon as possible is required. 1. A medicine reminder reminds a person to take their
medicine at the appointed time, which they had
Keywords—IOT, Smart reminder, Cost efficient, GSM previously pre-saved.
Module, GPS Module, Raspberry pi processor.
2. There is an SOS button that can be used in the event of
I. INTRODUCTION an emergency. Despite the fact that these two
The project is based on a multi-functional device that is characteristics existed previously, they were not used
highly advanced. The device has an inbuilt SOS button that in conjunction with an accident detection system. As a
can immediately notify pre-registered hospitals and local result, we've come up with this concepts [5].
guardians, and it can detect any type of collision and extract
the intensity of the collision or accident and above a certain II. LITERATURE SURVEY
intensity level, it notifies the victim's necessary contacts as Daniel Gloppestad Bajer [6] proposed that system was
well as the nearest hospital. The project is built on a created such that the phone is in charge of retrieving its own
combination of machine learning and IOT technology sensor readings as well as those from the watch, and
(Internet of Things). The need for Machine Learning evaluating the data to determine whether or not there has been
originates from the fact that a sensor must be able to a fall. Such analysis is based on three separate ways combined
distinguish between a normal jerk, hit, or movement and a real (two based on smart phone sensors data and one on the smart
collision. As a result, real accident collision patterns will be watch), as will be discussed later in the Fall Algorithm
necessary, and we will attempt to discover any device paragraph.
movement that is nearly identical to the actual collision
pattern. This trend must be followed in order to be successful. Joseph Santiago [7] proposed system is divided into two
When it comes to pattern matching, we'll need a Machine parts. The first is a pendant-like wearable gadget, and the
Learning Algorithm, and we'll try both KNN (K-Nearest second is a mobile application that runs on a smartphone.
Neighbour) and Support Vector Machine Algorithms to see Bluetooth will be used to communicate between these two
which one is the most efficient [1]. The device will be devices. For fall detection, the pendant contains a motion
sensor [8]. The necklace sends an alert to the cell phone when

978-1-6654-8316-2/22/$31.00 ©2022 IEEE

Authorized licensed use limited to: Jain University. Downloaded on June 14,2022 at 07:25:33 UTC from IEEE Xplore. Restrictions apply.
a person falls. The cell phone then phones and messages a STEP 3: If a match is detected between the data being
person of interest, telling them of the event's location. gathered and the pre-saved model, the user can select a
threshold for the level of match. If a match is found, the.
Cismas Alexandru [9], Proposed MPU6050 is an excellent
motion sensor; automobile sensors are designed to minimize STEP 4: In the event that another vehicle or pedestrian has
vibrations while also providing extremely precise readings. attempted to assist the victim by transporting him to another
The data fusion algorithm that uses acceleration, deceleration, nearby site or hospital, a continuous track of the victim's
and tilt angles has a high success rate [10]. There has never whereabouts is transmitted to loved ones and the hospital steps
been a false positive or a missed collision in any of the tests are carried out as shown in the below Figure 1.
conducted so far Crash detection is a crucial function for
motor cycle riders, and when combined with E-call. It can
reduce the time between the accident and the arrival of
emergency service. Nicky Kattukkaran [11] suggested system
will determine whether an accident has occurred and the
severity of the accident victim's/injuries. driver's Once a
serious accident has been determined, the system will look for
the nearest medical facility and alert them of the occurrence.
Because the correct location of the accident victim will be sent
via the accident victim's cell phone, [12] the rescue team will
be able to be on the scene quickly. Friends and relatives will
also receive messages from the system informing them about Fig. 1. System architecture
the occurrence.
S. Mohana Gowri [13] proposed an arduino uno
microcontroller is used to plan the proposed framework. The
hard-ware gadget will be placed near the vehicle's dashboard
in the suggested framework. An eye flicker sensor, vibration
sensor, beat rate sensor, GPS, GSM, buzzer, battery, switch,
power supply, and bell switch will be included in this gadget.
Rajvardhan Rishi [14] designed an AMS system will play an
important role. GSM technologies are utilized in situations
where the vehicle / police service must take rapid action to
reduce the severity. In the future, a dashboard will be
established to keep all accident information [15], which will
be accessible to friends, main road hospitals (rescue teams),
and government agencies. The AMS system will look up OBD
(On Board Diagnostic) data, which will make things easier. Fig. 2. Picture of LCD reminding about Time and Medicinena
In [16] author proposed when the detection threshold is
met, the notification system is activated. The GPS module
calculates the location of a vehicle accident, and then the GSM
module sends an accident message to the family members.
The latitude and longitude obtained from the GPS are sent to
the server via GPRS, and the server determines the nearest
hospital by using the [17] haversine formula to calculate the
great circle distance.
III. METHODOLOGY USED
STEP 1: Pre-saving data regarding medicine timings and
names, and when the time comes, the user is notified about his
or her medicine via a buzzer and an LCD screen as shown in
the Figure 2
STEP 2: On the device, there is an SOS button that is a little
them of the emergency and the user's less reachable to prevent
accidental pressing of the button. When pressed, the button
sends a text message to the user's phone using the SIM900A
GSM module, informing exact location as determined by the
NEO-6M GPS module. The device is equipped with
twoMPU6050 Accelerometers, which are positioned in two
separate orientations and continuously gather any vibrational
Fig. 3. Sample Emergency message received from the device
data that occurs to the device. In figure 3 the vibrational data
is constantly reviewed, compared, and matched against a pre-
saved Machine Learning Model of typical unintentional
vibrations.

Authorized licensed use limited to: Jain University. Downloaded on June 14,2022 at 07:25:33 UTC from IEEE Xplore. Restrictions apply.
 TABLE I. BASIC COMPONENTS USED IN THE PROPOSED METHOD
Raspberry With the addition of wireless
pi Zero W LAN and Bluetooth the
raspberry Pi Zero W has been
designed to be as flexible and
compact as possible with mini
connector and an unpopulated
40-pin GPIO

MPU605 MPU6050 is a Micro

0 Sensors Electro-mechanical
system(MEMS) it consists of
three-axis accelerometer and
three-axis gyroscope. It
helps us to measure velocity,
orientation, acceleration,
displacement and other
motion like features.

16X2 LCD Screen is used to

LCD display time, medicine
Screen names when reminded.

Ublox The Neo 6M GPS module

Fig. 4. Real time message received from the device. Situation is classified Neo 6M is a well performing
as a collision or accident. GPS complete GPS receiver
Locator with a built-in 25 x 25 x
4mm ceramic antenna,
In this case an SMS is sent to pre-saved contacts and the which provides a strong
city's featured hospitals as shown in the figure 3 & 4. satellite search capability.
With the power and signal
indicators, you can
monitor the status of the
module
SIM900A This module is a 2G based
GSM and over-the-network module.
GPRS This module can be used to
Module send messages, make calls
and turn on mobile hotspot.
This modules requires a SIM
card to be inserted.

V. CONCLUSION
In the everyday occupied work, monitoring the home
patient and overseeing their state of health ceaselessly is an
exceptionally troublesome task. People of all ages, especially
Fig. 5. Sample graph intensity matching using machine learning the elderly, should be checked and informed about their health
state on a regular basis in order to save their lives in life-
This graph represents the performance Metrics of the threatening situations. This project aims at developing a
proposed method. system that can detect the occurrence of the accident,
IV. TECHNICAL DESIGN reminding medications for chronic patients and an old age
person inform about the accident to the victim’s family on the
The following are basic hardware required to train and test pre- defined numbers and locates the nearest hospital and
the program (Table 1). inform them about the accident so they can immediately send
1. PyCharm/Jupyter IDE – For code writing and medical help. By efficiently reporting to the hospitals, this
developing approach has aided in lowering the time it takes to respond to
an accident. This was critical in saving the victims' lives.
2. Python 3.8 or later – Programming Language Therefore, we thought of building a single device which can
3. Kaggle – GPU Accelerated Environment for fast work as reminder machine for chronic patients, a SOS device
processing for aged, disable and needy people & detect any kind of
accident or collision. We also want to enhance the feature of
4. network>=2.3 library the device by giving the device a brain of its own which can
automatically sense for any collision and inform to both
5. numpy>1.17.1 library nearest hospital and local guardians via mail.
6. sci-kit library
7. sci-pi library
8. Torch-X libraries

Authorized licensed use limited to: Jain University. Downloaded on June 14,2022 at 07:25:33 UTC from IEEE Xplore. Restrictions apply.
 REFERENCES
[1] Md. Motaharul Islam, A. E. M Ridwan, Mekhala Mariam Mary2,Md
Fahim Siam, Sadia Anika Mumu, Shohag Rana.” Design and
Implementation of a Smart Bike Accident Detection System” In 2020
IEEE Region 10 Symposium (TENSYMP).
[2] Parag Parmar, Ashok M. Sapkal. “Real Time Detection and Reporting
of Vehicle Collision” in Conference on Trends in Electronics and
Informatics ICEI 2017.Detection”.International Conference on
Electronics.
[3] S.Mohana Gowri, P.Anitha, D.Srivaishnavi, M.Nithya” Internet of
Things based Accident Detection System”Third International
Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)
(I-SMAC 2019)IEEE Xplore Part Number:CFP19OSV-ART;
ISBN:978-1-7281- 4365-1
[4] Jacob, I. Jeena, and P. Ebby Darney. "Design of Deep Learning
Algorithm for IoT Application by Image based Recognition." Journal
of ISMAC 3, no. 03 (2021): 276-290.
[5] “Accident Detection Using Deep Learning”, 2020 2nd International
Conference on Advances in Computing, Communication Control and
Networking (ICACCCN), 18-19, Dec 2020.
[6] Rajvardhan Rishi, Sofiya Yede, Keshav Kunal, Nutan V. Bansode,
“Automatic Messaging System for Vehicle Tracking and Accident
Detection”, 2020 International Conference on Electronics and
Sustainable Communication Systems (ICESC), 2-4, July 2020.
[7] Ruhul Amin, Troyee Sharmistha Saha, Md. Faiyaz Bin Hassan,
Mahfuza Anjum, Md. Ishfak Tahmid, “loT Based Medical Assistant for
Efficient Monitoring of Patients in Response to COVID-19”, 2020 2nd
International Conference on Advanced Information and
Communication Technology (ICAICT), 28-29, Nov 2020.
[8] G. Karthy, M. Surya Kumar, Gudipadu Bhargav, K. Subramanyam,
“Medication Alerts and Supervisory of Health Using IOT”, 2021
Second International Conference on Electronics and Sustainable
Communication Systems (ICESC), 4-6, Aug 2021.
[9] Abhishek Madankar, Akhilesh Agrawal, Vedant Yede, “loT based
Advance Pill Reminder System for Distinct Patients”, 2021 Fifth
International Conference on I-SMAC (loT in Social, Mobile, Analytics
and Cloud) (I-SMAC), 11-13, Nov 2021.
[10] R. Lakshmi Devi, G. Saravanan, K. Sangeetha, S. Pavithra, S.
Thiyagarajan, “Smart Train Accident Detection And Prevention
System Using lot Technology”, 2021 International Conference on
System, Computation, Automation and Networking (ICSCAN), 30-31,
July 2020.
[11] Durgesh Kumar Yadav, Renu, Ankita, Iftisham Anjum, “Accident
Detection Using Deep Learning”, 2020 2nd International Conference
on Advances in Computing, Communication Control and Networking
(ICACCCN), 18-19, Dec 2020.
[12] Nazia Parveen, Ashif Ali, Aleem Ali, “IOT Based Automatic Vehicle
Accident Alert System”, 2020 IEEE 5th International Conference on
Computing Communication and Automation (ICCCA), 30-31, Oct
2020.
[13] Rajvardhan Rishi, Sofiya Yede, Keshav Kunal, Nutan V. Bansode,
“Automatic Messaging System for Vehicle Tracking and Accident
Detection”, 2020 International Conference on Electronics and
Sustainable Communication Systems (ICESC), 2-4, July 2020.
[14] Ruhul Amin, Troyee Sharmistha Saha, Md. Faiyaz Bin Hassan,
Mahfuza Anjum, Md. Ishfak Tahmid, “loT Based Medical Assistant for
Efficient Monitoring of Patients in Response to COVID-19”, 2020 2nd
International Conference on Advanced Information and
Communication Technology (ICAICT), 28-29, Nov 2020.
[15] Emergency System”, 2021 International Conference on
Communication information and Computing Technology (ICCICT),
25-27, June 2021
[16] G. Karthy, M. Surya Kumar, Gudipadu Bhargav, K. Subramanyam,
“Medication Alerts and Supervisory of Health Using IOT”, 2021
Second International Conference on Electronics and Sustainable
Communication Systems (ICESC), 4-6, Aug 2021.
[17] Abhishek Madankar, Akhilesh Agrawal, Vedant Yede, “loT based
Advance Pill Reminder System for Distinct Patients”, 2021 Fifth
International Conference on I-SMAC (loT in Social, Mobile, Analytics
and Cloud) (I-SMAC), 11-13, Nov 2021.

Authorized licensed use limited to: Jain University. Downloaded on June 14,2022 at 07:25:33 UTC from IEEE Xplore. Restrictions apply.
View publication stats