An Improved System For Brain Pathology Classification Using Hybrid Deep Learning Algorithm

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 34

An Improved System for Brain Pathology Classification using Hybrid Deep

Learning Algorithm
Presented By
P.UMARKHAN (MCA II Year)
Reg. No: 22091F0058
Under the esteemed guidance of
Dr. POGULA SREEDEVI
Associate Professor, Dept. of CSE

DEPARTMENT OF MASTER OF COMPUTER APPLICATIONS


RAJEEV GANDHI MEMORIAL COLLEGE OF ENGINEERING & TECHNOLOGY
(AUTONOMOUS)
NANDYAL-518501, (Estd-1995)
INDEX
 Abstract
 Introduction
 Existing System
 Proposed System
Software Design
Modules
Proposed Algorithms
SDLC
UML Diagrams
Software and Hardware Requirements
Abstract
With the progress of new generation wireless communication technology and machine learning
algorithms to deal with big data, a variety of smart systems are realized to bring comfort to human life.
Smart healthcare systems are one of the important developments recently.

• Such systems will become a necessary ingredient in our connected living. In this article, we propose a
new smart pathology detection system using deep learning. Sensors will capture electroencephalogram
(EEG) signals of a person. The classified decision of whether the signal belongs to a normal person or a
pathological person.

• The system is implemented the deep learning algorithm such as Convolutional Neural Network (CNN)
for classifying the input signal is pathology person or not effectively.

• Finally, the experimental results shows that some performance metrics such as accuracy, precision, recall
and f1 score.
Introduction

• Healthcare is one of the essential ingredients of human society. It is indispensable in our everyday life,
especially for elderly people, and people with chronic and psychological diseases. The management of these
diseases needs continuous and long-term human intervention.

• Physiological data need to be assessed frequently, and in a smart healthcare system, the data should be
transferred to and from a central server. Therefore, efficient communication between the stakeholders and the
data centers is of prime importance.

• In the case of emergency, the response should be in real time. A high response latency at data centers is a
potential risk in the healthcare system.

• Some other issues include packet drop during transmission and high computational complexity of data
manipulation.
Existing system
• In existing with the advancement of machine learning technologies, particularly deep learning, the
automated systems to assist human life are flourishing.

• In this paper, we propose an automatic electroencephalogram (EEG) pathology detection system based
on deep learning. Various types of pathologies can affect brain signals.

• Thus, the brain signals captured in the form of EEG signals can indicate whether a person suffers from
pathology or not.

• In the proposed system, the raw EEG signals are processed in the form of a spatio-temporal
representation. The spatio-temporal form of the EEG signals is the input to a convolutional neural
network (CNN).
Disadvantages

• It performed low accuracy.

• It does not efficient for large number of data’s.

• Theoretical limits.
Proposed system

• In proposed, the EEG signal dataset was collected from dataset repository.
Then, we can implement the preprocessing step.

• In this step, we can remove the unwanted noise from input signal by using
FIR (Finite Impulse Response).

• After that, we can split the dataset into test and train. Train data is used for
evaluation and test data is used for prediction. Then, we can implement the
deep learning algorithm such as CNN (Convolutional Neural Network).
Advantages

• High accuracy is performed when compared with existing.

• It is efficient for large number of dataset.

• The process is implemented with removing unwanted noise.


Flow Input Signal
diagram FIR

EEG Signal Preprocessing


Dataset

Data Splitting
Test

Train
CNN
Classification

Pathology

Prediction Normal

Accuracy

Performance
Specificity
Analysis
Sensitivity
Literature survey
Title Year Author Methodology Advantages Disadvantages

Edge cognitive 2018 Min Chen, This system is able to monitor  The Accuracy issues
computing based Yixue Hao and analyze the physical dynamic resource may come due to
smart healthcare health of users using cognitive allocation in handling such
system computing. It also adjusts the mechanism. massive data.
computing resource allocation 
of the whole edge computing Experimental
network comprehensively results show that
according to the health-risk the system
grade of each user. The realizes
experiments show that the optimization of
ECC-based healthcare system resources.
provides a better user
experience and optimizes the
computing resources
reasonably, as well as
significantly improving in the
survival rates of patients in a
sudden emergency.
Title Year Author Methodology Advantages Disadvantages

Emotion 2018 M. Shamim Simple signal preprocessing is  An  The


recognition using Hossain, performed in the edge clouds automatic audio- accuracy is very
secure edge and Ghulam and then the signals are visual emotion low.
cloud computing Muhammad transmitted to the core cloud. recognition
In the core cloud, a system based on
Convolutional Neural CNN
Network (CNN)-based, pre-  The
trained model is applied to audio-visual
extract deep-learned features features were
from image and speech fused using a
signals. cascaded deep
sparse
autoencoder.
Title Year Author Methodology Advantages Disadvantages

Lifelogging Data 2018 Po yang, . In LPAV-IoT, a methodology  Time  IoT may


Validation Model Zhikun Deng's specifying four layers and Through IoT, we reduce the cost for
for Internet of three modules is presented for can save a lot of diagnosis and
Things Enabled analyzing key factors time for diagnosis treatment for
Personalized impacting validity of and treatment. We patients, But the
Healthcare lifelogging physical activity. A can monitor cost of installing
series of validation rules are remotely rather all the devices and
designed with uncertainty than giving actual their maintenance
threshold parameters and physical visits. is quite high.
reliability indicators and
evaluated through
experimental investigations.
Following LPAV-IoT, a case
study on an IoT enabled
personalized healthcare
platform MHA connecting
three state-of-the-art wearable
devices and mobile apps are
carried out.
Title Year Author Methodology Advantages Disadvantages

Edge Caching 2019 Divya Gupta, The evaluations has been Tracking and Privacy of Data.
Based on Shalli Rani conducted to check the monitoring
Collaborative performance of the proposed Through the use
Filtering for strategy over various of IoT, we can
Heterogeneous benchmark strategies, such as keep track of data
ICN-IoT LCE, LCD, CL4M, and more efficiently.
Applications ProbCache. The analytical
results demonstrate the better
performance of our proposed
strategy with average gain of
15% for cache hit ratio, 12%
reduction in content retrieval
delay, and 28% reduced
average hop count in
comparison to best considered
LCD. We believe that the
proposed strategy will
contribute an effective
solution to the related studies
in this domain.
Title Year Author Methodology Advantages Disadvantages

Automated EEG 2021 Turker The Temple University  It is difficult to


signal Tuncer, Hospital (TUH) EEG dataset Automated whether to
classification Sengul Dogan is used to develop and abnormal EEG identify the signal
using chaotic evaluate our chaotic feature detection model is
local binary generation model. In this presented using
pattern work, the WPD is performed chaotic local
on EEG signals, and then binary pattern.
CLBP is applied on the
decomposed signals to extract
the features. The iterative
minimum redundancy
maximum relevancy
(ImRMR) is applied to select
the clinically significant
features. Finally, these
features are classified into
normal and abnormal EEG
classes using a support vector
machine (SVM) classifier.
Modules

• Input Signal

• Preprocessing

• Data splitting

• Classification

• Prediction

• Performance metrics
Modules Description
Input Signal

• Electroencephalography (EEG) is an efficient modality which helps to


acquire brain signals corresponds to various states from the scalp surface
area.

• These signals are generally categorized as delta, theta, alpha, beta and
gamma based on signal frequencies ranges from 0.1 Hz to more than 100
Hz.

• The data selection is the process of selecting the input signal for
classifying the pathology or not.
Input Signal
Preprocessing
• In signal processing, a finite impulse response (FIR) filter is a filter whose
impulse response (or response to any finite length input) is of finite
duration, because it settles to zero in finite time.

• We call the impulse response “finite” because there is no feedback loop in


this form of filter.

• If you put in an impulse as described earlier, zeroes will eventually be


output after the “1” valued sample has made its way in the delay line past all
the filter coefficients.
Preprocessing
Preprocessing
Data Splitting
• Data splitting is the act of partitioning available data into two portions, usually
for cross-validator purposes.

• One Portion of the data is used to develop a predictive model and the other to
evaluate the model's performance.

• Separating data into training and testing sets is an important part of evaluating
data mining models.

• Typically, when you separate a data set into a training set and testing set, most
of the data is used for training, and a smaller portion of the data is used for
testing.
Data Splitting
Classification

• In our process, we have to implement the deep learning algorithm such as


CNN.

• In 1D CNN, kernel moves in 1 direction. Input and output data of 1D


CNN is 2 dimensional. Mostly used on Time-Series data. In 2D CNN,
kernel moves in 2 directions.

• The 1D Convolution block represents a layer that can be used to detect


features in a vector.
Classification
Performance Metrics

• The Final Result will get generated based on the overall classification and
prediction. The performance of this proposed approach is evaluated using
some measures like,

• Accuracy

• Specificity

• Sensitivity
Prediction
• Here, we can classify or predict the input signal is pathology or not by using CNN.
Conclusion
• We conclude that, the input EEG signal dataset was taken from dataset repository. We are
developed the classification algorithms such as and deep learning algorithm like CNN.

• Finally, the result shows that some performance metrics such as Accuracy, specificity and
sensitivity.

• The application of the healthcare framework and some key technologies are studied in this article.
Several important edge and core functionalities are also figured out.

• As a case study, an EEG based pathology detection system is developed and examined.
Future Enhancement

• In the future, we should like to hybrid the two different machine learning.
In future, it is possible to provide extensions or modifications to the
proposed clustering and classification algorithms to achieve further
increased performance.

• Apart from the experimented combination of data mining techniques,


further combinations and other clustering algorithms can be used to
improve the detection accuracy.
References
• [1] M. Chen et al., “Edge Cognitive Computing Based Smart Healthcare
System,” Future Generation Computer Systems, vol. 86, 2018, pp. 403–11.

• [2] G. Muhammad et al., “Edge Computing with Cloud for Voice Disorders
Assessment and Treatment,” IEEE Commun. Mag., vol. 56, no. 4, Apr.
2018, pp. 60–65.

• [3] P. Yang et al., “Lifelogging Data Validation Model for Internet of Things
Enabled Personalized Healthcare,” IEEE Trans. Systems, Man, and
Cybernetics: Systems, vol. 48, no. 1, Jan. 2018, pp. 50–64.
References
• [5] Y. Hao et al., “Smart-Edge-CoCaCo: AI-Enabled Smart Edge with Joint Computation, Caching, and
Communication in Heterogeneous IoT,” IEEE Network, vol. 33, no. 2, Mar./Apr. 2019, pp. 58–64.

• [6] W. Shi et al., “Edge Computing: Vision and Challenges,” IEEE Internet of Things J., vol. 3, no. 5,
2016, pp. 637–64.

• [7] M. S. Hossain and G. Muhammad, “An Audio-Visual Emotion Recognition System Using Deep
Learning Fusion for Cognitive Wireless Framework,” IEEE Wireless Commun., vol. 26, no. 3, June 2019,
pp. 62–68.

• [8] M. A. Salahuddin, A. Al-Fuqaha, and M. Guizani, “Software-Defined Networking for RSU Clouds in
Support of the Internet of Vehicles,” IEEE Internet of Things J., vol. 2 no. 2, 2015, pp. 133–44.
HARDWARE REQUIREMENTS
system :Pentium 2.4 GHZ
Hard Disk :512
Ram :8 GB
SOFTWARE REQUIREMENTS
Operating system :windows 11
Language :python
Front End :anaconda navigator-spyder
THANK YOU

You might also like