RESEARCH PAPERS : SURVEY REPORT

1. Brain tumor prediction using naïve Bayes’ classifier and decision tree
algorithms : Danda Shashank Reddy1 , Chinta Naga Harshitha2 , Carmel Mary
Belinda

● The flock(group) of abnormal cells which are augmenting in the brain described
as a brain tumor. Irrespective of age, brain tumors are found to be present in any
person.
● Treatment of these various types of tumors can be easily done by several therapy
like chemotherapy, radiotherapy or sometimes combination therapy, but are
always life threatening. So, diagnosing brain tumors is very essential for carrying
the treatment forward.
● MRI has been proved as one of the most powerful and versatile brain imaging
modalities.
● The main objective is to
1.Estimate the occurrence of the tumor.
2.Identify the symptoms and the treatment necessary for the tumor.
● In this paper it is discussed about the data mining algorithms which are used to
predict the type of tumor that occurs based upon symptoms or treatment and from
the origination.
● Here we use mostly the decision tree algorithm and the naïve bayes classifier
algorithm.
● Decision Tree Algorithm :
1. Decision trees are the learning algorithms that construct a
classification tree to classify the data and it represents the visual
presentation of a problem .It decomposes a complex problem into
several sub problems with smaller size.
2. It is based on the approach of divide and conquers.The decision can be
made in a structured and formal approach. ( Refer fig1 ).
 ● Naïve Bayes’ Classifier :
1. The classifier known as Naïve Bayes Classifier is well-defined as the
collection of several algorithms that are classified based on Bayes’
Theorem.
2. The main basic Naïve Bayes hypothesis is known as each and every
feature varieties an:
▪ independent
▪ equal
contribution with a required outcome.
3. The classifier known as Naïve Bayes Classifier is well-defined as the
collection of several algorithms that are classified based on Bayes’
Theorem.
● Conclusion : Many data mining techniques are used for the detection of
brain tumor type among several kinds of patients. Using the above two algorithms
the type of tumor will been found and it allows analysis of historical data from
data sets which helps neurologists to predict the type of tumor.

2. DETECTION OF BRAIN TUMOR USING DATA SCIENCE: A SURVEY

Dr. Ganesh Khekare, Yashkumar Patel, Patel Nishit, Darsh Engineer, Prajapati Badal

● This article proposes a novel method to detect brain tumors from various MRI
images by deep learning method.
● It explains preprocessing techniques which are finalized for training and testing
and observing their impact on our MRI images dataset.
● MRI is considered as one of the most popular and it is widely used by the
doctors to detect the brain tumor. It is a very efficient technique because it
clearly differentiates between the brain structure and tissue on the basis of
contrast levels of the scan images.
● In this study we are going to implement it using machine learning techniques to
solve problems by providing brain tumor positive MRI scan image and negative
MRI scan image in the train class to train the system by machine learning so
that high accuracy can be achieved.
● Matrices are created by using MRI images with the help of MATLAB and then
PNN algorithm is used to classify the MRI images for the positive or negative
brain tumor detection.
● To make faster implementation of the R-CNN algorithm they used a deep
convolutional network. CNN algorithm and Chan-Vese algorithm are also used
for identifying meningioma and glioma brain tumors.
● The proposed flow chart is as shown in figure 1. It shows the process of
identification of tumors.

● This system trains the dataset of the MRI images which are having tumors and
even do not have any tumor both. with possible maximum precision.
● However, we would also like to conclude that unwanted behavior of the dataset
is not taken into consideration so MRI images dataset should be accurate to
achieve accurate results for the brain tumor detection. Both the machine
learning algorithms were quite efficient to detect the tumor part from the MRI
images.
● Conclusion : The survey of various brain detection systems using data science
and deep learning is done in this research article. The comparative analysis of
 training accuracy verses validation accuracy is done.( Refer fig3 )

3. Image Processing Techniques for Brain Tumor Detection: A Review, Vipin

Borole, Sunil S. Nimbhore , Dr. Seema S. Kawthekar

● MRI Imaging plays an important role in brain tumors for analysis, diagnosis and
treatment planning. It’s helpful for doctors to determine the previous steps of a
brain tumor.
● Brain tumors can be detected by benign or malignant types. The benign being
non-cancerous and malignant is cancerous. Malignant tumors are classified into
two types; primary and secondary tumor benign tumors are less harmful than
malignant.
● The MRI-Technique is most effective for brain tumor detection. In image
processing and image enhancement tools are used for medical image processing
to improve the quality of images.
● The various steps of MR imaging like; preprocessing, feature extraction,
 segmentation, post-processing, etc. which is used for finding the tumor area of
MRI-images. The figure-1 shows basic structure of feature extraction through
digital image processing.

● Image processing Techniques :

1. Median Filtering for Noise Removal

2. Various Denoising Filters
3. Image Enhancement
4. Edge Detection
5. Segmentation

● Conclusion : MRI images are best suitable for brain tumor detection. In this
study Digital Image Processing Techniques are important for brain tumor
detection by MRI images. The preprocessing techniques include different
methods like Filtering, Contrast enhancement, Edge detection is used for image
smoothing. The preprocessed images are used for post processing operations
like; threshold, histogram, segmentation and morphological, which is used to
enhance the images.