IRJET - Comparative Analysis of GUI based Prediction of Parkinson Disease using Machine Learning Approach

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 03 | Mar 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 2234
COMPARATIVE ANALYSIS OF GUI BASED PREDICTION OF PARKINSON
DISEASE USING MACHINE LEARNING APPROACH
Mr. S.RAMAKRISHNAN 1, R.SURYA2, R.VISHAL3 , M.SAMSON4, R.SHARATH KUMAR5
1Head Of The Department, Dept. of IT, Jeppiaar SRR Engineering College, Chennai, Tamil Nadu
2,3,4,5B.TECH., Dept. of IT, Jeppiaar SRR Engineering College, Chennai, Tamil Nadu
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Parkinson’s disease is the most prevalent
neurodegenerative disorder affecting more than 10 million
people worldwide. There is no single test which can be
administered for diagnosing Parkinson’s disease. Because of
these difficulties, to investigate a machine learning approach
to accurately diagnose Parkinson’s, using a given dataset. To
prevent this problem in medical sectors need to predict the
disease affected or not by finding accuracy calculation using
machine learning techniques. The aim is to research machine
learning based techniques for Parkinson disease byprediction
leads to best accuracy with finding classification report. The
analysis of dataset by supervised machine learning
technique(SMLT) to capture several information’s like,
variable identification, uni-variate analysis, bi-variate and
multi-variate analysis, missing value treatments and analyze
the data validation, data cleaning/preparing and data
visualization are going to be done on the whole given dataset.
To propose, a machine learning-based method to accurately
predict the disease by speech symptom by prediction resultsin
the form of best accuracy and additionally compare the
performance of various machine learningalgorithmsfrom the
given hospital dataset with evaluation classification report,
identify the result shows that GUI with best accuracy with
precision, Recall ,F1 Score specificity and sensitivity.
1.INTRODUCTION
Machine learning is to predict the future from past data.
Machine learning focuses on the development of Computer
Programs that can change when exposedto newdata andthe
basics of Machine Learning, implementation of a simple
machine learning algorithm usingpython.Itfeedthetraining
data to an algorithm, and the algorithm uses this training
data to give predictions on a new test data. Machinelearning
can be roughly separated in to three categories. There are
supervised learning, unsupervised learning and
reinforcement learning. Supervisedlearningprogramis both
given the input data and the corresponding labeling to learn
data has to be labeled by a human being beforehand.
Unsupervised learning is no labels. It provided to the
learning algorithm. This algorithm has to figure out the
clustering of the input data. Finally, Reinforcement learning
dynamically interacts with its environment and it receives
positive or negative feedback to improve its
performance.Data scientists use many different kinds of
machine learning algorithms to discover patterns in python
that lead to actionable insights. Classification predictive
modeling is the task of approximating a mapping function
from input variables(X) to discrete output variables(y).This
data set may simply be bi-class (like identifying whether the
person is male or female or that the mail is spam or non-
spam) or it may be multi-class too.
1.1 What is Supervised Machine Learning?
Supervised Machine Learningisthatthemajorityofpractical
machine learning uses supervised learning. Supervised
learning is where have input variables (X) and an output
variable (y) and use an algorithm to find out the mapping
function from the input to the output is y = f(X). The goal is
to approximate the mapping function so well that once you
have new input file (X) that you simply can
predict the output variables (y) for that data. Techniques of
Supervised Machine Learning algorithms include logistic
regression, multi-class classification, Decision Trees and
support vector machines etc. Supervised learning requires
that the info wont to train the algorithm is already labeled
with correct answers. Supervised learning problems are
often further grouped into Classification problems. This
problem has as goal the constructionofa succinctmodel that
can predict the value of the dependent attribute from the
attribute variables. The difference between the two tasks is
the fact that the dependent attribute is numerical for
categorical for classification. Given one or more inputs a
classification model will attempt to predict the worth of 1 or
more outcomes. A classification problem is when the output
variable may be a category, like “red” or “blue”.
1.2 What is Parkinson’s disease
Parkinson’s disease a long term degenerativedisorderof the
central nervous system that affects the motor control of a
patient by affecting predominately dopamine producing
neurons in a specific area of the brain. The main problem in
detecting the disease timely is the visible symptoms appear
mostly at the later stage where cure no longer becomes
possible. There is no correct reason proved yet that results
to cause of Parkinson’s, hence scientists are still conducting
extensive research to find out its exact cause. Though some
abnormal genes that become prominent due to elderly

appear to lead to Parkinson’s in some people but there is no
evidence to proof this. Though there are a couple of
procedures early Parkinson’s detection. Dopamine
transporter single-photon emission computed tomography
can be used to effectively diagnosis Parkinson’s by detecting
amount of dopamine deficiency in the concerned patient’s
brain cell at a considerable early stage. Parkinson’s disease
(PD) affects approximately one million Americans and can
cause several motor and non-motor symptoms. One of the
secondary motor symptoms that people with PD may
experience is change in speech, or speech difficulty. Not
everyone with PD experiences same symptoms, and not all
patients will have changes in their speech.
2. MODULES DESCRIPTION
2.1 VARIABLE IDENTIFICATION PROCESS AND
DATA VALIDATION PROCESS
Validation techniques in machinelearningare usedtogetthe
error rateof the MachineLearning (ML) model, which canbe
considered as close to the true error rate of the dataset. If the
info volume is large enough to be representative of the
population, you'll not need the validation techniques.
However, in real-world scenarios, to work with samples of
data that may not be a true representative of the population
of given dataset. To finding the missingvalue,duplicatevalue
and description of data type whether it is float variable or
integer. The sample of knowledge wont to provide an
unbiased evaluation of a model fit on the training dataset
while tuning model hyper parameters. The validation set is
employed to guage a given model, but this is often for
frequent evaluation. It as machine learning engineers uses
this data to fine-tune the model hyper parameters. Data
collection, data analysis, and therefore the process of
addressing data content, quality, and structurecan add up to
a time-consuming to-do list.Duringthemethodofknowledge
identification, it helps to know your data and its properties;
this data will assist you choose which algorithm to use to
create your model. For example, time series data can be
analyzed by regression algorithms; classification algorithms
can be used to analyze discrete data. (For example to show
the data type format of given dataset)
2.1.1 DATA VALIDATION/CLEANING/PREPARING
PROCESS
Importing the library packages with loading given dataset.
To analyzing the variable identification by data shape, data
type and evaluating the missing values, duplicate values. A
validation dataset is a sample of data heldback fromtraining
your model that is used to give an estimate of model skill
while tuning model's and procedures that you can use to
form the simplest use of validation and test datasets when
evaluating your models. Data cleaning / preparing by
rename the given dataset and drop the column etc. to
analyze the uni-variate, bi-variate andmulti-variateprocess.
The primary goal of knowledge cleaning is todetectandtake
away errors and anomalies to extend the worth of
knowledge in analytics and deciding .
2.1.2 DATA PREPROCESSING
Data Preprocessingmaybeatechniquethat'swonttoconvert
the data into a clean data set. In other words, whenever the
info is gathered from different sources it's collected in raw
format which isn't feasible for the analysis. To achieving
better results from the applied model in Machine Learning
method of the data has to be in a proper manner. Some
specified MachineLearning model needs information during
a specified format; for instance , Random Forest algorithm
doesn't support null values. Therefore, to execute random
forest algorithm null values need to be managed from the
first data set. And another aspect is that data set should be
formatted in such how that quite one Machine Learning and
Deep Learning algorithms are executed in given dataset.
2.2 DATA VISUALIZATION
Data visualization is an important skill in applied statistics
and machine learning. Statistics does indeed specialise in
quantitative descriptions and estimations of knowledge .
Data visualization provides an important suite of tools for
gaining a qualitative understanding. This can be helpful
when exploring and going to know a dataset and may help
with identifying patterns, corrupt data, outliers, and far
more. With a little domain knowledge, data visualizations
can be used to express and demonstrate key relationshipsin
plots and charts that are more visceral and stakeholders
than measures of association or significance. Data
visualization and exploratory data analysis are whole fields
themselves and it will recommend a deeper dive into some
the books mentioned at the end.
2.3 LOGISTIC REGRESSION
It is a statistical procedure for analyzing a knowledge set
during which there are one or more independent variables
that determine an outcome. The goal of logistic regressionis
to seek out the simplest fitting model to explain the
connection between the dichotomous characteristic of
interest (dependent variable = response or outcome
variable) and a group of independent (predictor or
explanatory) variables.Logisticregressionmaybea Machine
Learning classification algorithm that's wont to predict the
probability of a categorical variable . In logistic regression,
the variable may be a binary variable that contains data
coded as 1 (yes, success, etc.) or 0 (no, failure, etc.).
2.3.1 DECISION TREE
Decision tree builds classification or regression models
within the sort of a tree structure. It breaks down a

knowledge set into smaller and smaller subsets while at an
equivalent time an associated decision tree is incrementally
developed. Decision tree builds classification or regression
models within the sort of a tree structure. Each time a rule is
learned, the tuples covered by the principles are
removed.This process is sustained on the training set until
meeting a termination condition. It is constructed in a top-
down recursive divide-and-conquer manner. All the
attributes should be categorical. Otherwise, they should be
discretized in advance. Attributes within the top of the tree
have more impact towards within the classification and that
they are identified using the knowledge gain concept. A
decision tree are often easilyover-fittedgeneratingtoomany
branches and should reflect anomalies thanks to noise or
outliers.
2.4 SUPPORT VECTOR MACHINES
A classifier that categorizes the info set by settinganoptimal
hyper plane between data. I chose this classifier becauseit is
incredibly versatile within the number of various kernelling
functions which will be applied and this model can yield a
high predictability rate. Support Vector Machines are
perhaps one among the foremost popular and talked about
machine learning algorithms. They were extremely popular
round the time they were developed within the 1990s and
still be the go-to method for a high-performing algorithm
with little tuning.
2.4.1 RANDOM FOREST
Random forest may be a sort of supervisedmachinelearning
algorithm supported ensemble learning. Ensemble learning
may be a sort of learning where you join differing types of
algorithms or same algorithm multipletimestomakea more
powerful prediction model. The random forest algorithm
combines multiple algorithm of an equivalent type i.e.
multiple decision trees, leading to a forest of trees,hencethe
name "Random Forest". The random forest algorithm are
often used for both regression and classification tasks.
2.5 K-NEAREST NEIGHBOR
K-Nearest Neighbor is a supervised machine learning
algorithm which stores all instances correspond to training
data points in n-dimensional space.Thek-nearest-neighbors
algorithm may be a classification algorithm, and it's
supervised: it takes a bunch of labeled points and uses them
to find out the way to label other points. To label a
replacement point, it's at the labeled points closest thereto
new point (those are its nearest neighbors), and has those
neighbors vote, so whichever label themostofthe neighbors
have is that the label for the new point (the “k” is that the
number of neighbors it checks). Makespredictionsaboutthe
validation set using the entire training set. KNN makes a
prediction about a new instance by searching through the
entire set to find the k “closest” instances.
2.5.1 NAIVE BAYES ALGORITHM
The Naive Bayes algorithm is an intuitive method that uses
the possibilities of every attribute belongingto everyclassto
form a prediction. It is the supervised learning approach
you'd come up with if you wanted to model a predictive
modeling problem probabilistically. This is a strong
assumption but results in a fast and effective method. The
probability of a category value given a worth of an attribute
is named the contingent probability . By multiplying the
conditional probabilities together for every attribute for a
given class value, we've a probability of a knowledge
instance belonging thereto class. To make a prediction we
will calculate probabilities of the instancebelongingto every
class and choose the category value with the very best
probability.Naive Bayes may be a statistical classification
technique supported BayesTheorem.Itisoneofthesimplest
supervised learning algorithms. NaiveBayesclassifieristhat
the fast, accurate and reliable algorithm. Naive Bayes
classifier assumes that the effect of a particular feature in a
class is independent of other features. For example, a loan
applicant is desirable or not depending on his/her income,
previous loan and transaction history, age, and location.
Even if these features are interdependent, thesefeatures are
still considered independently. This assumption simplifies
computation, and that's why it is considered as naive. This
assumption is called class conditional independence.
2.6 GRAPHICAL USER INTERFACE
Tkinter is a python library for developing GUI (Graphical
User Interfaces). We use the tkinter library for creating an
application of UI (User Interface), to create windows and all
other graphical user interface and Tkinter will come with
Python as a standard package, it can be used for security
purpose of each users or accountants. There will be two
kinds of pages like registration user purpose and loginentry
purpose of users.It is important tocomparethe performance
of multiple different machine learning algorithms
consistently and it will discover to create a test harness to
compare multiple different machine learning algorithms in
Python with scikit-learn. It can use this test harness as a
template on your own machine learning problems and add
more and different algorithms to compare.

2.6.1 ACCURACY CACULATION
Table-1:Performance measurements of ML algorithm for
speech
Table-2: Performance measurements of ML algorithm for
tremor
Parameters LR DT
Precision 1 1
Recall 0.80 1
F1-Score 0.89 1
Sensitivity 0.8 1
Specificity 1 1
Accuracy (%) 95.83 100
Table-3: Performance measurements confusion matrix for
speech
Parameters LR DT
TP 39 40
TN 11 13
FP 4 2
FN 5 4
TPR 0.88 0.90
TNR 0.73 0.86
FPR 0.26 0.13
FNR 0.11 0.09
PPV 0.90 0.95
NPV 0.98 0.76
Table-4: Performance measurements confusion
matrix for tremor
Parameters LR DT
TP 19 19
TN 4 5
FP 1 0
FN 0 0
TPR 1 1
TNR 0.8 1
FPR 0.2 0
FNR 0 0
PPV 0.95 1
NPV 1 1
Parameters LR DT
Precision 0.69 0.76
Recall 0.73 0.87
F1-Score 0.71 0.81
Sensitivity 0.73 0.866
Specificity 0.88 0.90
Accuracy (%) 84.74 89.83

3. SYSTEM TECHNIQUES
Design is meaningful engineering representation of
something that's to be built. Software design is a process
design is the perfect way to accurately translate
requirements in to a finished software product. Design
creates a representation or model, provides detail about
software arrangement, architecture, interfaces and
components that are necessary to implement a system.
SYSTEM ARCHITECTURE
Fig-1:Architecture Diagram
HARDWARE REQUIREMENTS
 Processor : i3/i4
 Hard disk : minimum 300 GB
 RAM : minimum 4 GB
SOFTWARE REQUIREMENTS
Operating System : Windows
Tool : Anaconda with Jupyter
Notebook
SNAPSHOT
Fig-2:Parkinson disease prediction for tremor symptom
Fig-3: Parkinson disease prediction for speech symptom
ADVANTAGES
The scope of this project is to investigate a dataset of
Parkinson normal person records and patient records for
medical field using machinelearningtechnique.Toanalyzing
the prediction of Parkinson disease person is moreaccuracy
with comparing algorithm and try to reducethedoctor’srisk
of factor behind detecting the patient. Easy to predicting the
diagnose Parkinson with doctors can detecting the patient
testing result time is reduced
APPLICATION
A hospital wants to automate the diagnosis of patient (real
time) based on the patient detail provided while filling
online/ offline application form. To automate this process,
they have given a problem to identify the patient segments,
those are have detecting disease to list to show doctors.
FUTURE ENHANCEMENT
Hospitals want to automate the detecting the disease
persons from eligibility process (real time) based on the
account detail.To automate this process by show the
prediction result in web application or desktop
application.To optimize the work to implement in Artificial
Intelligence environment.
CONCLUSION
The analytical process started from data cleaning and
processing, missing value, exploratory analysis and finally
model building and evaluation. The best accuracy on public
test set is higher accuracy score will be finding out. This
brings some of the following insights about diagnose the
Parkinson disease. Early diagnosis of Parkinson’s is most
important for the patient to reduce its impact. It presenteda
prediction model with the aid of artificial intelligence to
improve over human accuracy and provide withthescopeof
early detection. It can be inferred from this model that, area
analysis and use of machine learning technique is useful in
developing prediction models that can help a doctor reduce
the long process of diagnosis anderadicateanyhumanerror.

REFERENCES
[1] X. Feng et al., “A language-independent neural network
for event detection,” Sci. China Inf. Sci., vol. 61, no. 9, pp. 92–
106, 2018
[2] F. Karim et al., “LSTM fully convolutional networks for
time series classification,” IEEE Access, vol. 6, pp. 1662–
1669, 2018.
[3] W. Liu et al., “Learning efficient spatial-temporal gait
features with deep learning for human identification,”
Neuroinformatics, vol. 16, pp. 457– 471, 2018.
[4] S. Yeung et al., “Every moment counts: Dense detailed
labeling of actions in complex videos,”Int.J.Comput.Vis.,vol.
126, no. 2–4, pp. 375–389, 2018.
[5] M. M. Hassan et al., “A robust human activity recognition
system using smartphonesensorsanddeeplearning,”Future
Gener. Comput. Syst., vol. 81, pp. 307–313, 2018.
.

IRJET - Comparative Analysis of GUI based Prediction of Parkinson Disease using Machine Learning Approach

More Related Content

What's hot (18)

Similar to IRJET - Comparative Analysis of GUI based Prediction of Parkinson Disease using Machine Learning Approach (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET - Comparative Analysis of GUI based Prediction of Parkinson Disease using Machine Learning Approach