Eating Habit and Health Monitoring System using Android Based Machine Learning

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 03 | Mar -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 2141
Eating Habit and Health Monitoring System Using Android Based
Machine Learning
Avinash Palve1,Snehal More2, Shivani Chaudhari3, Akshay Katke4, Kartike Kampassi5
1Professor, Avinash Palve, Dept. of Computer Engineering, TCOER, Pune, Maharashtra, India.
2Snehal More, Department of Computer Engineering, TCOER, Pune, Maharashtra, India
3Shivani Chaudhari, Department of Computer Engineering, TCOER, Pune, Maharashtra, India
4Akshay Katke, Department of Computer Engineering, TCOER, Pune, Maharashtra, India
5Kartike Kampassi, Department of Computer Engineering, TCOER, Pune, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – Now a days, there are many threats to human
health. Nutrition-related diseases are now becoming a
dangerous threat to human health. Balancing energy intake
and expenditure is crucial steptomaintainhealthylifestyle. To
collect the acoustic signals during eating and chewing, a
wearable device is to be worn around the person's neck . An
embedded hardware prototype will collect the food intake
data. Signals collected will be processedbythehardware. blue
tooth signal will send to Smartphone where food types are
recognized. We use hidden Markov models to detect chewing
or swallowing events. Developed an application on the
Smartphone, which verifies the food intake results and gives
relevant suggestions on healthier habits. Itnotifiestheuser by
giving notification to the user.
Key Words: Health, Signal, Food, Smart phone,
Acoustic, Bluetooth etc..
1.INTRODUCTION
It is Important to maintain a healthy lifestyle in our day to
day life. Daily food intake must be worth neither it may lead
to various diseases and other consequences. If proper care
isn't taken then severe disorders may become threat.
Present solutions to measure calorie expenditure are, such
as Fit-bit, Philips Direct Life, etc. However, continuouslyand
non-invasive monitoring on calorie intake remains a
challenge. Currently present solutions rely on self-reports,
which are not convenient because the food intakes differ
from person to person or to different age groups.Todevelop
easy-to-use solution to detect the daily food intake and its
expenditure endorse Android Based Healthcare System
Using Machine Learning. We propose this system as it is
desirable to develop more accurate and more easy methods
to monitor the food intake. It is user friendly, easy to use,
interactive with any smart phone which can handle it and
understand its working in detail properly.
2. RELATED WORK
Advances in wearable devices continues to play a major role
in improving human healthcare as it provides several
opportunities which help to balance food intake and energy
consumption. Advancement in wearable devices which help
to monitor body organs,movementandaction.Manysensors
are being deployed in the system to collect the health
related data of user. Interpretation of data plays the major
role. Real-world degradation of signals are more prone to
noise which lowers down the quality of systemperformance
in speech recognition systems.HiddenMarkovModel isused
for construction of speech components which are spectral
whose performance is degraded by unwanted noise. At each
step there is discrete state among the number of possible
states which are finite. In the frequency domain encoding of
speech parameters is done by number of speech analysis.
3. PROPOSED SYSTEM
Main goal of this Technology is to make user aware of
his/her daily food intake. Provide the suggestions according
to it. Provide customized suggestionstoindividual regarding
healthier eating habits. Develop an application food
recognition results information in a most user-friendly way.
Design of an Health Monitoring System which can be
interfaced with an android Smartphone. The embedded
system give updatestheuserbyusingSmartphoneregarding
its health.
3.1 Body Mass Index and Calorie Intake
To measure the daily calorie intake of a person weuseone of
the most easiest method. The method is to measure your
body mass index (BMI). The BMI is a calculation of your
weight in relationship to your height.
 Calculating your BMI
1. Measure and record your height(in cm).
2. Measure your weight(in kg).
3. Calculate your BMI by using formula:
BMI = Weight (kg)________ / Height (m2) ________ =
__________
Result will be either Underweight Normal Overweight
Obese.

Fig1-BMI Chart
3.2 Calories In vs. Calories Out
Your energy balance is the relationship between “energy in”
(food calories taken in into the blood through food and
drink) and “energy out” (calories being used in our daily
activities).The same amount of ENERGY INTAKE (calories
consumed) and ENERGY OUT(calories burned) over the
time = weight remains the proportionally same.
More IN than OUT over time = gain in weight .
More OUT than IN over time = loss of weight .
3.3Calculating Your Energy Needs
Fig2-Harris-Benedict equation
We use the Harris-Benedict equation to calculate the energy
needs of a person.
4.SYSTEM ARCHITECTURE
Fig3-Hardware Board
Android based health monitoring system is mainly
comprises of two components: An embedded unit for
acoustic data gathering and pre-processing, retrieval and
smart-phone unit run the application to give suggestions to
the user. These notifications are based on the calorie
consumption by the user. Embedded unit interact with the
smart-phone via Bluetooth connection. Architecture is
divided into two system are given below:
1)Embedded System
1)Vibration Sensor -The wearabledevicewornaround neck
is used for sensing acoustic signals .2)ADC, 3)AVR Micro
controller : ATmega32 (40 pins). , 4) Bluetooth : HC-05.
2)Smartphone Unit
The application developed on the Smartphone works as a
data manager and provides an interface to the individual.To
start the application when the user starts to eat, the system
will store the detailed data into a database at smart phone
side. The user can not only check the records, but also go
through the notifications on healthier eating habits which
are obtained by analyzing the data collected.
 WORKING PROCESS
The wearable device is worn around neck while eating food.
Vibration sensor converts vibrations from the skin to
acoustic signals. Embedded system is used for data
processing and transmission. When the acoustic data
collected from the throat vibration sensor. They are

amplified and filtered to achieve better signal quality. Then
the analog signals are converted into digital signal for later
steps. The digital signals are passed to the microcontroller
for segmentation and later for framing. The microcontroller
is also responsible for frame admission control of raw
signals from the vibration sensor. The data frames are being
sent to the Bluetooth module through UART using SPI
transport protocol and further sent to the smart phone by
the Bluetooth module with a SPP profile.
 WORKING ON SMARTPHONE:
Step1-Open the application
Step2-Login or Sign Up For new account
After user log in into application. His profile get created
which contain his/her weight, height, gender, age.
Step3-Calorie gain and burn calculation window
Step4-Train the system

Step5-Create and Save the patterns by training
system
Step6-Display result and suggestions
3. CONCLUSIONS
Android based health monitoringsystemiscomposedoftwo
components: an embedded unit for acoustic data gathering,
retrieval as well as per-processing, andSmartphoneunitrun
the application to give suggestions to the user for healthier
eating habits. These suggestions are based on the calorie
consumption by the individual. Embedded unit interacts
with the Smartphone unit via Bluetooth connectivity.
4.FUTURE SCOPE
Concept of image processing can be used to recognize the
type of food intake. Solids can be differentiated from other
solids and liquids can be differentiated from other liquids.
The system can also be supported by use of audio
processing. In audio processing ,user needs to call out the
name of the food item and it acts according to it.
ACKNOWLEDGEMENT
The research is on the project topic “Android Based Health
Monitoring System Using Machine”. We are grateful to our
esteemed guide, Prof. Avinash Palve, for providing the right
guidance and advice at the crucial times. Last but not the
least; we would like to thank our friends and family for the
support and encouragement they have given us during this
entire course of our work.
REFERENCES
[1]By Yin Bi, Min song Lv, Chen Song, Wenyao Xu, Nan Guan,
Member, IEEE, and Wang Yi, Fellow, IEEE. Auto-dietary- A
Wearable Acoustic Sensor System for Food Intake
Recognition in Daily Life
[2]International Journal of InnovativeResearchinAdvanced
Engineering (IJIRAE) [1] Issue 2, Volume 2 (February 2015)
ISSN: 2349-2163 www.ijirae.com Hidden Markov Model
Based Robust Speech Recognition Vikas Mulik Vikram Mane
Imran Jamadar
[3]By E. S. Sazonov, O. Makeyev, S. Schuckers, P. Lopez-
Meyer, E. L. Melanson, and M. R. Neuman, “Automatic
detection of swallowing events by acoustical means for
applications of monitoring of ingestive behavior,” IEEE
Trans. Biomed. Eng., vol. 57, no. 3, pp. 626–633, Mar. 2010
[4]IEEE Pervasive Computing, vol. 8, no. 2, pp. 62–70,
Apr./Jun. 2009.O. Amft and G. Troster, “On-body sensing
solutions for automatic dietary monitoring,”
[5] Physiol. Meas.,vol. 29, 2008.“Non-invasive monitoring of
chewing and swallow-ing for objective quantification of
ingestive behavior,”E.S. Sazonov et al.
[6]Hmm-based acousticevent detection with adaboost
feature selectionsMing Liu Hao Tang Mark Hasegawa-

Johnson Xi Zhou,Xiaodan Zhuang and Thomas Huang.
MultimodalTechnologiesfor PerceptionofHumans,Springer
BerlinHeidelberg, 2007.
[7]“A comparative study on feature selection in text
categorization,” By Y. Yang and J. O. Pedersen, in Proc. 14th
Int. Conf. Mach. Learn, 1997.

Eating Habit and Health Monitoring System using Android Based Machine Learning

More Related Content

What's hot (17)

Similar to Eating Habit and Health Monitoring System using Android Based Machine Learning (20)

More from IRJET Journal (20)

Recently uploaded (20)

Eating Habit and Health Monitoring System using Android Based Machine Learning