Similarity Report ID: oid:16158:55543985

PAPER NAME AUTHOR

NTCC REPORT 010424.docx Megha Pachuri

WORD COUNT CHARACTER COUNT

4794 Words 29646 Characters

PAGE COUNT FILE SIZE

24 Pages 790.7KB

SUBMISSION DATE REPORT DATE

Apr 1, 2024 9:26 PM GMT+5:30 Apr 1, 2024 9:27 PM GMT+5:30

15% Overall Similarity

The combined total of all matches, including overlapping sources, for each database.
15% Internet database 14% Publications database
Crossref database Crossref Posted Content database
2% Submitted Works database

Excluded from Similarity Report

Bibliographic material Quoted material
Cited material Small Matches (Less then 14 words)
Manually excluded sources

Summary
NTCC REPORT

INTRODUCTION

BRAIN TUMOR DETECTION SYSTEM

The human body consists of many organs, and the most important and important organ among them is
the brain. One of the most common causes of brain dysfunction is stroke. Tumors are nothing more th
an cells growing uncontrollably. Brain tumors grow to absorb all the nutrients that healthy cells and ti
ssues need, causing the brain to fail. Currently, doctors manually look at MRI images of the patient's b
rain to find the location and area of the brain tumor. This results in false tumor detection and is consid
ered time consuming. Brain cancer is a3serious disease that kills many people. Techniques are availabl
e to detect and isolate brain tumors for early diagnosis. Classifying cancer is the most difficult task in
diagnosis. This study involves a system that uses a computerbased service to detect tumors and uses a
neural network algorithm to classify tumors on MRI images of different patients. Various types of ima
ge processing, such as image segmentation, image enhancement and object removal, are used for brai
n diagnosis in MR images of cancer patients. There are four steps to identifying a brain tumor using i
mage processing tools: image preprocessing, image segmentation, feature extraction, and classificatio
n. Image processing and neural network systems are used to increase performance in MR images, and
neural network systems are used to improve brain function and classification.

OVERVIEW OF BRAIN AND BRAIN TUMOR

The human brain, situated in the head and protected by the skull, is the central component of the human
nervous system. Its main role is to regulate and coordinate all bodily functions. As an organ, it enables
humans to adapt and withstand various environmental stimuli. Through the brain, humans are able to
perform actions, express thoughts, and experience emotions. This section will provide an overview of
the brain's structure to aid in comprehending its fundamental functions.

.
There are two main types of brain tumors: primary and secondary. Primary tumors, also known as
benign tumors, are slow-growing cells in the brain, specifically gliomas which originate from
astrocytes. While these tumors are less aggressive, they can still put pressure on the brain and impair
its function. On the other hand, secondary tumors, or malignant tumors, are more aggressive and spread
quickly to other tissues. They originate from other parts of the body and can be caused by metastatic
cancer cells from areas like the brain, lungs, or kidneys. These tumors are highly malignant and often
result from other cancers such as lung, kidney, or bladder cancer. There are two main types of brain
tumors: primary and secondary. Primary tumors, also known as benign tumors, are slow-growing cells
in the brain, specifically gliomas which originate from astrocytes. While these tumors are less
aggressive, they can still put pressure on the brain and impair its function. On the other hand, secondary
tumors, or malignant tumors, are more aggressive and spread quickly to other tissues. They originate
from other parts of the body and can be caused by metastatic cancer cells from areas like the brain,
lungs, or kidneys. These tumors are highly malignant and often result from other cancers such as lung,
kidney, or bladder cancer.

.
MAGNETIC RESONANCE IMAGING (MRI)
In 1969, Raymond v. Damadian was credited with creating the initial magnetic image. By 1977, the first
MRI images for the human body were developed, solidifying it as the most advanced technique available.
Thanks to MRI technology, we now have the ability to view intricate details of internal brain structure and
identify various types of human tissue. In comparison to other medical imaging methods such as X-rays and
CT scans, MRI images offer superior quality. This advancement has been especially beneficial in detecting
brain tumors in the human body. There are various types of MRI images used for mapping tumor-related
changes, such as T1 weighted, T2 weighted, and FLAIR (Fluid attenuated inversion recovery).

APPLICATION
The primary goals of these application is to identify ,tumors efficiently, enabling prompt treatment to
saves lives at risk. this tool benefits both doctor and patient by offering a faster, more accurate
alternative to manual identification. The application was developed to address the limitation of manual
identification, providing a more effitient and precise solution for users.

OBJECTIVE
The primary objective of these application is to detect tumor accuretley .the main motivation for
creating this application is to ensure timely treatment and safeguard lives at risk. this tool is beneficial
for both medical professionals and patients. manual identification is slower , less precise and less
efficient compared to this designed application ,which aims to overcome these challenges.

RELATED WORK:

DEEP LEARNING BASED TECHNIQUE:

.
B. Srikanth et al. presented 16-layer VGG-16 deep neural network that
process enhanced images from a preprocessing phase through convolutional
layerto extract features and downsample them.their method significantly
improved the accuracy of multiclassifying brain tumor mr-images,achieving
a 98% accuracy after 20 training iterations.
2
GS Tandel et al. developed five clinical multiclass dataset and utilized a
transfer learning based convolutional neural network(cnn) to enhanced
brain tumor classification using MRI images. Their cnn model
outperformed six other machine learning classification approaches,
achieving high accuracy rates across different cross validation procedures.

Another study proposed a CNN technique for three-class brain tumor

classification, distinguishing between glioma, meningioma and pituatry
tumor.by laveraging pretained google net for feature extraction and proven
based classification, they achieved an average classification accuracy of
98%,surpassing exsisting method.

In a different approach, reserchers presented a deep inception residual

network for three- class brain tumor classification , enhancing the accuracy
of classification . their model, tested on a public brain tumor imagin dataset
, outperformed state-of-the-art techniques with an accuracy exceeding 99%.

Using transfer learning another study developed a brain tumor classification

method with MRI images , employing VGG16,RESNET50,DENSNET
AND VGG19 network to detect common brain cancers. Their model
significantly improved classification performance compared to another
methods.

A novel RCNN architecture was designed for brain tumor classification

,utilizing a low complexity framework to reduce processing time. This
approche achieved high accuracy in identifying tumor area in MRI dataset
of glioma , meningioma and pituitary tumors.

Furthermore , imagenet-based vision transformer(vit) model were trained

and fine-tuned for brain tumor classifiacation , achieving high accuracy
rates on a three-class brain tumor dataset. The vit models surpassed

.
individual algoritham performance, when ensembled, demonstrating
superior accuracy in classification task.

Machine Learning-Based Techniques

Pareek and colleagues introduced a method for detecting and classifying brain tumor using T1-weighted
MRI brain imaging. Their approach involved a supervised classification process with principal
component analysis for feature extraction,. The experiment showed that KSVM achieved 97% accuracy
in tumor classification. Another novel method focused on enhancing MRI quality and feature selection
through normalization, powerful feature extraction, and histogram of gradient techniques, using a
support vector machine for classification. This method achieved 90.27% accuracy, surpassing, previous
techniques. Additionally , a quantum fully self neural network (QFS-Net) utilizing quantum
correlations for brain lesion segmentation was proposed ,employing a quatrit-based counter-
propagating technique for improved network performance.

2.3. Hybrid-Based Techniques

2
Khairandish and colleagues introduced a hybrid approach that combines CNN and SVM with threshold
segmentation for brain tumor classification.. their proposed CNNA and SVM hybrid achieved and
increased overall accuracy of 98.4959%. pre-trained network like AlexNet, GoogLeNet , shuffleNet
and ResNet -18 were utilized to extract feature from tumor region and adjacent tissues. TO address
potential loss of low-level tumor data, a shallow network was incorporated al;ongside deep features,
and SVM and KNN classifiers were trained using merged feature sets . experiment result showed that
data expension and ROI expension significantly improved average sensitivity by approximately
11.72% highlighting the importance of surrounding tissues data. The fusion of deep and shallow
features compensated for missing low level information , with the deep feature extraction process
carried out using Resnet-18, showcasing competitive results with state-of-the-art technique.

.
 —

3. Brain Tumor Classification Model Based CNN (BCM-CNN)

The proposed brain tumor classoification model based on CNN(BCM-CNN) utilizes an advanced
convolutional neural network model for brain tumor diagnosis. This model involves hyperparameter

.
optimization followed by training with inseption-ResNetV2. THE output of the model is
binary(0:normal 1:tumor) and lavrages common pre-trained model like inception- RESNETV2 enhance
the accuracy of brain tumor diagnosis.

3.1. CNN Hyperparameters Optimization

The subsection deleves into specific hyperperameters used in the study. configuration hyperparameters
are external variable not inherent to the model and cannot be deducted from the data. These
hyperperameters fall into two primary categories: those defining the network structure and those
utilized during network training. Tables 2 details the hyperperameters explore in this research.

ADSCFGWO for CNN Hyperparameters

The ADSCFGWO algorithm is an optimization techniques that combines the strenghthes of the sine
cosine and grey wolf algorithm to enhance performance in CNN hypermeter optimization. This
algorithm divides the potential solution population into exploration and exploitation groups to
efficiently search for optimal solutions.by balancing mining and exploration, it effectively avoids local
optima and explores the search space accurately. The key features of ADSCFGWO include maintaining
control over the balance between the two groups and using a dynamic mechanism to prevent stagnation
in the search space. The algorithm identifies the most suitable solutions marked as S alpha , s beta , s

.
gama and updated the position during the search process based on prey position using specific
calculations outlined in the algorithm.

.
2
3.3. 3D U-Net Architecture Segmentation Model

The 3D U-Net architechture is a model used for rapid and precise image segmentation feauturing both
anexpanding and contracting track.the contaracting path follows a standard convolutional network
design with repeated 3*3 unpaded convolutional layers, RElu activation, 2*2 max pooling for
downsampling. The expending path involves up-sampling, a 2*2 convolution layer, combination with
a cropped feature map, and additional convolution layers with RELU activation. The model comprises
23 convolutional layers and effitiently segments images due to its speed, taking less than a second for
a 512*512 image on a modern GPU.

The 3D U- Net an extension of the U-Net model, is particularly significant in medical image
segmentation. It enhances the original U-net by incorporating 3D covolution, max pooling, and
deconvolution blocks for feature extraction and recovery from volumetric data. Batch normalization is
integrated to accelerate convergence and prevent bottlenecks. The encoder extract features from input
images, while the decoder generates segmented masks by minimizing the cost function. This model
segmentation process involves dividing the dataset into training validation, and testing sets.

4.1. Dataset Description

.
The reserch articles provided focus on brain tumor detection and classification using deep learning
techniques applied to MRI images. One study introduces a brain tumor classification model(BCM-
CNN) that enhances CNN capabilities through hyperperameters optimization employing the
ADSCFGWO algorithm. Another study proposes a hybrid deep CNN model for multi classification of
brain tumor images, creating three CNN models for classifying different types of brain tumors with grid
search optimization of key hyperperameters. These models aim to classify MRI images into different
categories of brain tumors such as benign,malignant, meningioma, pituitary and metastatic tumors.

Furthermore , an improved arithmetic optimization algorithm with forced switching mechanism is

introduced to enhance the searching quality of the arithmetic optimization algorithm (AOA a ) for better
exploration and exploitation in global optimization problems. The enhanced algorithm, called AOA,
integrates forced switching mechnasim to help search agents avoid local optima effectively.
Experimental results show the superiority of iAOA over other comparative algorithms in various tests
functions and training problems.

Data augmentation technique specifically image augmentation , are employed in study to generate
new training data by altering exsisting data. In image augmentation , the input dataset undergoes
various transformations like horizontal and vartical shifts, flips, rotation and zooms. Shift
augmentation shift all pixels of an MRI images horizontally or vartically while maintaining the same
dimensions. Horizontal or vertical flips reverse the rows and columns of pixels. Rotation
augmentation randomly rotates MRI images between 0 and 360 degree clockwise. This process helps
create diverse training data for deep learning models without changing the original data set.

2
Performance Metrics Used in CNN:
A conventional computer aided diagnostic approach can be evaluated using various key performance
metrices, such as accuracy, precision, f1-score, recall, specificity and sensitivity. True positive indicates
the number of correctly identified defective cases, while false positive represent cases incorrectly
labelled as defective. False negative referse to cases incorrectly classified as benign and true negative
denotes cases correctly identified as benign. These metrices plays a crucial role in assesseing the
performance of diagnostic model.

.
The BCM-CNN Evaluation:

Table 6 present a comparison of the effectiveness .of the proposed approach(BCM-CNN) with
4
commonly used classifiers such as CNN, decision tree(dt).., linear discriminant(LD), support vector
machine (svm ) and k-nearest neighbors(KNN). The assessments is conducted using the default
parametrs foreach classifiers to evalute their performance.

In the classification study, the BCM-CNN model, enhanced by the ADSCFGWO algorithm,
significantly improved the performance of the CNN by adjusting its hyperperameters. When utilized as
2
a classifiers, the BCM-CNN model outperformed other classifiers with an accuracy of 0.99980004,
2
sensitivity of 0.99980004, specificity of 0.99980004, precision of , negative predictive value of
0.99980004 and F1-SCORE of 0.9998. following the SVM linear model with an accuracy of
0.968992248, the KNN model with an accuracy of 0.96525090965 and the LD model with an accuracy
of 0.961538462, the simple CNN model achieved the second highest accuracy at 0.9765625. . in
contrast the DT model had the lowest accuracy at 0.956022945 due to the proposed method enhancing
the overall performance of the CNN,.
Table 7 provide the statistical overview of the BCM-CNN based classifiers and compared it with other
classifiers based on 11 runs and 80 iterations of the ADSCFGWO algorithm to ensure the stability of
the proposed method. Table 8 display the test result of the compared and proposed classifiers using a
one way anova test, while table 9 discusses the test results using the Wilcoxon signed rank test,
indicating a significant difference between the results of the BCF-CNN classifier and other classifier

.
.
.
2
The accuracy of the proposed BCM-CNN and other comparative methods is depicted in the
box plot in image This visualization showcase the maximum accuracy outcomes achieved by
the optimization algorithm based on BCM-CNN. The accuracy histogram for the algorithm
presented and compared within the Bin center range of 0.946 to 1.0, demonstrates the stability
and effectiveness of the suggested algorithm.

The box plot illustrates the the accuracy of the proposed BCM-CNN method
and its comparison with other method. This visual representation showcase
the highest accuracy achieved by the optimization algorithm utilizing BCM-
CNN. Additionally, the accuracy histogram focusing on the bin center range

.
 1
of 0.946 to 1.0, demonstrate the stability and effectiveness of the proposed
algorithm.

Figure 5

Box plot of accuracy for the BCM-CNN model under consideration and the contrasting
models.

Histogram of accuracy for the BCM-CNN model under consideration and the
contrasting models.

The figure shows the residual, QQ (quantile-quantile), homoscedasticity plots and heat map for the
proposed and compared techniques. Possible problems can be observed in the residuals and plots against
the plot of the original data set. The independent variable is plotted on the horizontal axis, while the
residual values are plotted on the vertical axis. The ideal situation is achieved if the residual values are
scattered randomly and uniformly along the horizontal axis. The residual value is calculated as follows
when both the mean and the sum of the residuals equal zero: (Actual-Predicted Values). The figure
shows a plot of residuals. Plotting patterns in the residual plot can be used to determine whether a model

.
 1
is linear or non-linear and which one is best. The projected score for the dependent variable is examined
visually along with homogeneity of variance or heteroskedasticity. When the error term, also known as
noise or random disturbance in the relationship between the dependent and independent variables, is
constant across all values of the independent variables, this situation is referred to as homoscedasticity.
The heteroskedasticity graph shown in the figure improves the accuracy of the research results. Any
offense can be detected quickly and easily.

1
The accuracy of the proposed BCM-CNN and the comparison methods is shown in the box plot in
Figure 5. This plot shows the maximum accuracy results that the optimization algorithm based on BCM-
CNN was able to produce. Based on the number of values with the Bin Center range (0.946–1.0), the
accuracy histogram for the algorithms that were presented and compared is shown in Figure 6 , which
confirms the stability of the proposed algorithm.

Figure 5
Box plot of accuracy for the BCM-CNN model under consideration and the contrasting models.

Figure 6

.
Histogram of accuracy for the BCM-CNN model under consideration and the contrasting
models.

Figure 7 display the residuals, QQ(CHECK-CHECK), similarity graph, and heat map of the proposed
and compared method. It highlight potential issues in the residual values and plot compared to the
original data plot. The residual graph templete aids in determining whether the model is linear and non
linear and which is more suitable.by visually inspecting the predicted scores of the dependent variable.
One can assess the homeogenity or hyterogenity of variance. Homoscedasticity, where the error term
remains constant across all independent variable values, enhances study accuracy. Detecting any
violations promptly and effortlessly is crucial for improving the study precision.

A QQ PLOT depicted in the figure , is a type of probability plot that compares two probability
distribution by plotting quantiles against each other.in the qq plot, the distribution of point aligning

with the line indicates a good fit between the distributions. The linear relationship between the actual

and predicted residuals in the qq plot signifies the effectiveness of the proposed techniques. This

visualization tools also showcase heat maps for both the proposed and compared algorithms providing

insight into their performance and highlighting any differences between them.

3D U-Net Segmentation Model:-

.
 2
The segmentation process in brain tumor detection involves classes such as not tumor, non enhancing
tumor (red color), EDEMA (green color) and ENHANCING(yellow color) which were later combined
into three classes for efficiency. A 3D U-Net segmentation model was utilized with the BRAts 2021
daataset, split into 70% for training, 20% for validation , and 10% for testing. This model achieved a
segmentation validation accuracy of 99.33% and a validation loss of 0.01, demonstrating high accuracy
in brain tumor detection compared to existing techniques in classification and segmentation.

.
Conclusions and Future Work:
Numerous research have extensively explored various algorithms to accurately and effectively detect
and classify brain tumor . the utilization of deep learning(DL) and a pretrained convolutional neural
network(CNN) model for analyzing medical images particularly for categorizing brain tumor , has
become viable. The primary objective of this study was to create a model using theBRAts 2021Task 1
dataset, achieving an impressive accuracy of 99.99% .however a significant drawback of the proposed
algorithm is its prolonged processing time due to additional optimization steps. Furthermore, its
practicality is limited by the size of the training data. To overcome this limitation, future efforts will
concentrate on expanding the dataset to improve generalization. Additionally upcoming research aims
to integrates predictive capabilities alongside classification.

REFERENCES:-

1. Al-Galal S.A.Y., Alshaikhli I.F.T., Abdulrazzaq M.M. MRI brain tumor medical
images analysis using deep learning techniques: A systematic review. Health
Technol. 2021;11:267–282. doi: 10.1007/s12553-020-00514-6. [CrossRef] [Google
Scholar]
2. Rahman M.L., Reza A.W., Shabuj S.I. An internet of things-based automatic brain
tumor detection system. Indones. J. Electr. Eng. Comput. Sci. 2022;25:214–222.
doi: 10.11591/ijeecs.v25.i1.pp214-222. [CrossRef] [Google Scholar]
3. Key Statistics for Brain and Spinal Cord Tumors. [(accessed on 20 September 2022)].
Available online: https://www.cancer.org/cancer/brain-spinal-cord-tumors-
adults/about/key-statistics.html
4. Ayadi W., Elhamzi W., Charfi I., Atri M. Deep CNN for Brain Tumor
Classification. Neural Process. Lett. 2021;53:671–700. doi: 10.1007/s11063-020-
10398-2. [CrossRef] [Google Scholar]
5. Liu J., Li M., Wang J., Wu F., Liu T., Pan Y. A survey of MRI-based brain tumor
segmentation methods. Tsinghua Sci. Technol. 2014;19:578–595.
doi: 10.1109/tst.2014.6961028. [CrossRef] [Google Scholar]
6. Amin J., Sharif M., Haldorai A., Yasmin M., Nayak R.S. Brain tumor detection and
classification using machine learning: A comprehensive survey. Complex Intell.
Syst. 2021;8:3161–3183. doi: 10.1007/s40747-021-00563-y. [CrossRef] [Google
Scholar]
7. Jayade S., Ingole D.T., Ingole M.D. Review of Brain Tumor Detection Concept using
MRI Images; Proceedings of the 2019 International Conference on Innovative Trends
and Advances in Engineering and Technology (ICITAET); Shegoaon, India. 27–28
December 2019; [CrossRef] [Google Scholar]
8. Yang Y., Yan L.F., Zhang X., Han Y., Nan H.Y., Hu Y.C., Hu B., Yan S.L., Zhang
J., Cheng D.L., et al. Glioma Grading on Conventional MR Images: A Deep Learning

.
Study With Transfer Learning. Front. Neurosci. 2018;12:804.
doi: 10.3389/fnins.2018.00804. [PMC free article] [PubMed] [CrossRef] [Google
Scholar]
9. Nazir M., Shakil S., Khurshid K. Role of deep learning in brain tumor detection and
classification (2015 to 2020): A review. Comput. Med. Imaging
Graph. 2021;91:101940. doi: 10.1016/j.compmedimag.2021.101940. [PubMed]
[CrossRef] [Google Scholar]
10. El-Kenawy E.S.M., Mirjalili S., Abdelhamid A.A., Ibrahim A., Khodadadi N., Eid
M.M. Meta-Heuristic Optimization and Keystroke Dynamics for Authentication of
Smartphone Users. Mathematics. 2022;10:2912.
doi: 10.3390/math10162912. [CrossRef] [Google Scholar]
11. El-kenawy E.S.M., Albalawi F., Ward S.A., Ghoneim S.S.M., Eid M.M.,
Abdelhamid A.A., Bailek N., Ibrahim A. Feature Selection and Classification of
Transformer Faults Based on Novel Meta-Heuristic
Algorithm. Mathematics. 2022;10:3144.
doi: 10.3390/math10173144. [CrossRef] [Google Scholar]
12. El-Kenawy E.S.M., Mirjalili S., Alassery F., Zhang Y.D., Eid M.M., El-Mashad
S.Y., Aloyaydi B.A., Ibrahim A., Abdelhamid A.A. Novel Meta-Heuristic Algorithm
for Feature Selection, Unconstrained Functions and Engineering Problems. IEEE
Access. 2022;10:40536–40555.
doi: 10.1109/ACCESS.2022.3166901. [CrossRef] [Google Scholar]
13. Ibrahim A., Mirjalili S., El-Said M., Ghoneim S.S.M., Al-Harthi M.M., Ibrahim
T.F., El-Kenawy E.S.M. Wind Speed Ensemble Forecasting Based on Deep Learning
Using Adaptive Dynamic Optimization Algorithm. IEEE Access. 2021;9:125787–
125804. doi: 10.1109/ACCESS.2021.3111408. [CrossRef] [Google Scholar]
14. El-kenawy E.S.M., Abutarboush H.F., Mohamed A.W., Ibrahim A. Advance
Artificial Intelligence Technique for Designing Double T-shaped Monopole
Antenna. Comput. Mater. Contin. 2021;69:2983–2995.
doi: 10.32604/cmc.2021.019114. [CrossRef] [Google Scholar]
15. Samee N.A., El-Kenawy E.S.M., Atteia G., Jamjoom M.M., Ibrahim A.,
Abdelhamid A.A., El-Attar N.E., Gaber T., Slowik A., Shams M.Y. Metaheuristic
Optimization Through Deep Learning Classification of COVID-19 in Chest X-Ray
Images. Comput. Mater. Contin. 2022;73:4193–4210.
doi: 10.32604/cmc.2022.031147. [CrossRef] [Google Scholar]
16. Lee G., Nho K., Kang B., Sohn K.A., Kim D. Predicting Alzheimer’s disease
progression using multi-modal deep learning approach. Sci. Rep. 2019;9:1952.
doi: 10.1038/s41598-018-37769-z. [PMC free article] [PubMed] [CrossRef] [Google
Scholar]
17. Sharma K., Kaur A., Gujral S. Brain Tumor Detection based on Machine Learning
Algorithms. Int. J. Comput. Appl. 2014;103:7–11. doi: 10.5120/18036-
6883. [CrossRef] [Google Scholar]

.
18. Agrawal M., Jain V. Prediction of Breast Cancer based on Various Medical
Symptoms Using Machine Learning Algorithms; Proceedings of the 2022 6th
International Conference on Trends in Electronics and Informatics (ICOEI);
Tirunelveli, India. 28–30 April 2022; pp. 1242–1245. [CrossRef] [Google Scholar]
19. Rabbi M.F., Mahedy Hasan S.M., Champa A.I., AsifZaman M., Hasan M.K.
Prediction of Liver Disorders using Machine Learning Algorithms: A Comparative
Study; Proceedings of the 2020 2nd International Conference on Advanced Information
and Communication Technology (ICAICT); Dhaka, Bangladesh. 28–29 November
2020; pp. 111–116. [CrossRef] [Google Scholar]
20. Swain D., Pani S.K., Swain D. A Metaphoric Investigation on Prediction of Heart
Disease using Machine Learning; Proceedings of the 2018 International Conference on
Advanced Computation and Telecommunication (ICACAT); Bhopal, India. 28–29
December 2018; pp. 1–6. [CrossRef] [Google Scholar]
21. Zhu W., Sun L., Huang J., Han L., Zhang D. Dual Attention Multi-Instance Deep
Learning for Alzheimer’s Disease Diagnosis With Structural MRI. IEEE Trans. Med
Imaging. 2021;40:2354–2366. doi: 10.1109/TMI.2021.3077079. [PubMed]
[CrossRef] [Google Scholar]
22. Abdelhamid A.A., El-Kenawy E.S.M., Alotaibi B., Amer G.M., Abdelkader M.Y.,
Ibrahim A., Eid M.M. Robust Speech Emotion Recognition Using CNN+LSTM Based
on Stochastic Fractal Search Optimization Algorithm. IEEE Access. 2022;10:49265–
49284. doi: 10.1109/ACCESS.2022.3172954. [CrossRef] [Google Scholar]
23. Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Erhan D.,
Vanhoucke V., Rabinovich A. Going deeper with convolutions; Proceedings of the
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Boston,
MA, USA. 7–12 June 2015; [CrossRef] [Google Scholar]
24. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition;
Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition
(CVPR); Las Vegas, NV, USA. 27–30 June 2016; [CrossRef] [Google Scholar]
25. Alhussan A.A., Khafaga D.S., El-Kenawy E.S.M., Ibrahim A., Eid M.M.,
Abdelhamid A.A. Pothole and Plain Road Classification Using Adaptive Mutation
Dipper Throated Optimization and Transfer Learning for Self Driving Cars. IEEE
Access. 2022;10:84188–84211.
doi: 10.1109/ACCESS.2022.3196660. [CrossRef] [Google Scholar]
26. Srikanth B., Suryanarayana S.V. Multi-Class classification of brain tumor images
using data augmentation with deep neural network. Mater. Today Proc. 2021
doi: 10.1016/j.matpr.2021.01.601. [CrossRef] [Google Scholar]
27. Tandel G.S., Balestrieri A., Jujaray T., Khanna N.N., Saba L., Suri J.S. Multiclass
magnetic resonance imaging brain tumor classification using artificial intelligence
paradigm. Comput. Biol. Med. 2020;122:103804.
doi: 10.1016/j.compbiomed.2020.103804. [PubMed] [CrossRef] [Google Scholar]

.
28. Deepak S., Ameer P. Brain tumor classification using deep CNN features via
transfer learning. Comput. Biol. Med. 2019;111:103345.
doi: 10.1016/j.compbiomed.2019.103345. [PubMed] [CrossRef] [Google Scholar]
29. Kokkalla S., Kakarla J., Venkateswarlu I.B., Singh M. Three-class brain tumor
classification using deep dense inception residual network. Soft
Comput. 2021;25:8721–8729. doi: 10.1007/s00500-021-05748-8. [PMC free
article] [PubMed] [CrossRef] [Google Scholar]
30. Özlem P., Güngen C. Classification of brain tumors from MR images using deep
transfer learning. J. Supercomput. 2021;77:7236–7252. doi: 10.1007/s11227-020-
03572-9. [CrossRef] [Google Scholar]
31. Kesav N., Jibukumar M. Efficient and low complex architecture for detection and
classification of Brain Tumor using RCNN with Two Channel CNN. J. King Saud
Univ.-Comput. Inf. Sci. 2022;34:6229–6242.
doi: 10.1016/j.jksuci.2021.05.008. [CrossRef] [Google Scholar]
32. Tummala S., Kadry S., Bukhari S.A.C., Rauf H.T. Classification of Brain Tumor
from Magnetic Resonance Imaging Using Vision Transformers Ensembling. Curr.
Oncol. 2022;29:7498–7511. doi: 10.3390/curroncol29100590. [PMC free
article] [PubMed] [CrossRef] [Google Scholar]
33. Pareek M., Jha C.K., Mukherjee S. Advances in Intelligent Systems and
Computing. Springer; Singapore: 2020. Brain Tumor Classification from MRI Images
and Calculation of Tumor Area; pp. 73–83. [CrossRef] [Google Scholar]
34. Ayadi W., Charfi I., Elhamzi W., Atri M. Brain tumor classification based on hybrid
approach. Vis. Comput. 2020;38:107–117. doi: 10.1007/s00371-020-02005-
1. [CrossRef] [Google Scholar]
35. Konar D., Bhattacharyya S., Panigrahi B.K., Behrman E.C. Qutrit-Inspired Fully
Self-Supervised Shallow Quantum Learning Network for Brain Tumor
Segmentation. IEEE Trans. Neural Netw. Learn. Syst. 2022;33:6331–6345.
doi: 10.1109/TNNLS.2021.3077188. [PubMed] [CrossRef] [Google Scholar]
36. Khairandish M., Sharma M., Jain V., Chatterjee J., Jhanjhi N. A Hybrid CNN-SVM
Threshold Segmentation Approach for Tumor Detection and Classification of MRI
Brain Images. IRBM. 2022;43:290–299.
doi: 10.1016/j.irbm.2021.06.003. [CrossRef] [Google Scholar]
37. Öksüz C., Urhan O., Güllü M.K. Brain tumor classification using the fused features
extracted from expanded tumor region. Biomed. Signal Process.
Control. 2022;72:103356. doi: 10.1016/j.bspc.2021.103356. [CrossRef] [Google
Scholar]
38. Khan M.A., Ashraf I., Alhaisoni M., Damaševičius R., Scherer R., Rehman A.,
Bukhari S.A.C. Multimodal Brain Tumor Classification Using Deep Learning and
Robust Feature Selection: A Machine Learning Application for
Radiologists. Diagnostics. 2020;10:565. doi: 10.3390/diagnostics10080565. [PMC
free article] [PubMed] [CrossRef] [Google Scholar]

.
39. Kadry S., Nam Y., Rauf H.T., Rajinikanth V., Lawal I.A. Automated Detection of
Brain Abnormality using Deep-Learning-Scheme: A Study; Proceedings of the 2021
Seventh International conference on Bio Signals, Images, and Instrumentation
(ICBSII); Chennai, India. 25–27 March 2021; [CrossRef] [Google Scholar]
40. Irmak E. Multi-Classification of Brain Tumor MRI Images Using Deep
Convolutional Neural Network with Fully Optimized Framework. Iran. J. Sci. Technol.
Trans. Electr. Eng. 2021;45:1015–1036. doi: 10.1007/s40998-021-00426-
9. [CrossRef] [Google Scholar]
41. Khafaga D.S., Alhussan A.A., El-Kenawy E.S.M., Ibrahim A., Eid M.M.,
Abdelhamid A.A. Solving Optimization Problems of Metamaterial and Double T-Shape
Antennas Using Advanced Meta-Heuristics Algorithms. IEEE Access. 2022;10:74449–
74471. doi: 10.1109/ACCESS.2022.3190508. [CrossRef] [Google Scholar]
42. Ronneberger O., Fischer P., Brox T. Lecture Notes in Computer Science. Springer
International Publishing; Berlin, Germany: 2015. U-Net: Convolutional Networks for
Biomedical Image Segmentation; pp. 234–241. [CrossRef] [Google Scholar]
43. Özgün Ç., Abdulkadir A., Lienkamp S.S., Brox T., Ronneberger O. Medical Image
Computing and Computer-Assisted Intervention–MICCAI 2016. Springer International
Publishing; Berlin, Germany: 2016. 3D U-Net: Learning Dense Volumetric
Segmentation from Sparse Annotation; pp. 424–432. [CrossRef] [Google Scholar]
44. BRaTS 2021 Task 1 Dataset, RSNA-ASNR-MICCAI Brain Tumor Segmentation
(BraTS) Challenge 2021. [(accessed on 20 September 2022)]. Available
online: https://www.kaggle.com/datasets/dschettler8845/brats-2021-
task1?select=BraTS2021_Training_Data.tar
45. Nour M., Cömert Z., Polat K. A Novel Medical Diagnosis model for COVID-19
infection detection based on Deep Features and Bayesian Optimization. Appl. Soft
Comput. 2020;97:106580. doi: 10.1016/j.asoc.2020.106580. [PMC free
article] [PubMed] [CrossRef] [Google Scholar]
46. Tharwat A., Gaber T., Ibrahim A., Hassanien A.E. Linear discriminant analysis: A
detailed tutorial. AI Commun. 2017;30:169–190. doi: 10.3233/AIC-
170729. [CrossRef] [Google Scholar]
47. Benmahamed Y., Teguar M., Boubakeur A. Application of SVM and KNN to Duval
Pentagon 1 for transformer oil diagnosis. IEEE Trans. Dielectr. Electr.
Insul. 2017;24:3443–3451. doi: 10.1109/TDEI.2017.006841. [CrossRef] [Google
Scholar]
48. Benmahamed Y., Kemari Y., Teguar M., Boubakeur A. Diagnosis of Power
Transformer Oil Using KNN and Nave Bayes Classifiers; Proceedings of the 2018 IEEE
2nd International Conference on Dielectrics (ICD); Budapest, Hungary. 1–5 July 2018;
[CrossRef] [Google Scholar]

.
.
 Similarity Report ID: oid:16158:55543985

15% Overall Similarity

Top sources found in the following databases:
15% Internet database 14% Publications database
Crossref database Crossref Posted Content database
2% Submitted Works database

TOP SOURCES
The sources with the highest number of matches within the submission. Overlapping sources will not be
displayed.

Hanaa ZainEldin, Samah A. Gamel, El-Sayed M. El-Kenawy, Amal H. Alh...

1 9%
Crossref

mdpi.com
2 5%
Internet

Soundarya C, Kalaiselvi A, Surya J. "Brain Tumor Detection Using Imag...

3 <1%
Crossref

doctorpenguin.com
4 <1%
Internet

Sources overview
 Similarity Report ID: oid:16158:55543985

Excluded from Similarity Report

Bibliographic material Quoted material
Cited material Small Matches (Less then 14 words)
Manually excluded sources

EXCLUDED SOURCES

researchgate.net
13%
Internet

Jacobs University, Bremen on 2023-05-31

<1%
Submitted works

link.springer.com
<1%
Internet

Excluded from Similarity Report