Problem Statement

Brain and other nervous system cancer is the 10th leading cause of death for men and women.
Brain cancer affects all age groups, and experts estimate that roughly 700,000 Americans currently
live with a brain tumor. In 2021, more than 84,000 people were diagnosed with primary brain
tumors.

Abstract
A Brain tumor is considered as one of the aggressive diseases, among children and adults. Brain
tumors account for 85 to 90 percent of all primary Central Nervous System(CNS) tumors. Every
year, around 11,700 people are diagnosed with a brain tumor. The 5-year survival rate for people
with a cancerous brain or CNS tumor is approximately 34 percent for men and36 percent for
women. Brain Tumors are classified as: Benign Tumor, Malignant Tumor, Pituitary Tumor, etc.
Proper treatment, planning, and accurate diagnostics should be implemented to improve the life
expectancy of the patients. The best technique to detect brain tumors is Magnetic Resonance
Imaging (MRI). A huge amount of image data is generated through the scans. These images are
examined by the radiologist. A manual examination can be error-prone due to the level of
complexities involved in brain tumors and their properties.
Application of automated classification techniques using Machine Learning(ML) and Artificial
Intelligence(AI)has consistently shown higher accuracy than manual classification. Hence,
proposing a system performing detection and classification by using Deep Learning Algorithms
using ConvolutionNeural Network (CNN), Artificial Neural Network (ANN), and
TransferLearning (TL) would be helpful to doctors all around the world.

Objectives
● Develop an ML-based system for early and accurate detection of brain tumor.
● Create a user-friendly web application to analyze the predicted results.
● Utilize image processing and machine learning to analyze brain MRIs.
● Implement a robust and efficient classification model.
● Train the model on a diverse dataset of brain MRIs.
● Develop a responsive and engaging user interface.
● Continuously update and improve the system for better prediction.

1
Introduction:
Early and accurate detection of brain tumor is crucial for better treatment and increased patient
survival chances. However, traditional brain tumor detection methods can be time-consuming,
expensive, and unreliable. Recent advancements in image processing and machine learning offer
new solutions, enabling automated brain tumor detection.
In this project, we leverage these technologies to build a system for brain tumor detection. Our
approach is based on CNN network architectures to analyse brain MRIs to classify images based
on visual disease symptoms. By providing instant and accurate diagnoses, our project leads to early
and better treatments, higher survival rates, and reduced costs.

Scope:
Detecting brain tumors using Convolutional Neural Networks (CNN) has a significant scope in
medical image analysis and healthcare. Here are various points that highlight the scope of your
brain tumor detection project:

1) Medical Diagnosis Improvement: Brain tumor detection can significantly improve the early
diagnosis of brain tumors, leading to better treatment outcomes and increased patient survival
rates.

2) Automation: Automation through CNNs can reduce the reliance on human radiologists and
enhance the efficiency of medical image analysis, especially in regions with a shortage of medical
professionals.

3) Early Detection: Early detection of brain tumors can lead to more effective treatments and a
higher chance of complete recovery. CNNs can help in identifying tumors at an early stage.

4) Reduction in Misdiagnosis: Machine learning models can reduce the risk of misdiagnosis, which
can have severe consequences in the case of brain tumors.

5) Cost-Efficiency: Automating tumor detection can reduce overall healthcare costs associated with
manual image analysis.

6) Scalability: The project can be scaled to analyze a larger dataset, incorporating more diverse
cases and improving the model's accuracy.

2
Literature Survey:
Brain tumors are a significant health concern, and their early detection is crucial for timely
treatment and improved patient outcomes. In recent years, there has been significant research in the
field of brain tumor detection, with a focus on developing accurate and efficient methods using
various imaging modalities and machine learning techniques.

1. Medical Imaging Modalities for Brain Tumor Detection:

a. Magnetic Resonance Imaging (MRI):
- Review studies on the use of MRI in brain tumor detection.
- Explore the evolution of MRI techniques for better tumor visualization.

b. Computed Tomography (CT) Imaging:

- Examine the role of CT scans in detecting brain tumors.
- Discuss the advantages and limitations of CT compared to MRI.

2. Machine Learning and Deep Learning Approaches:

a. Traditional Machine Learning Methods:
- Summarize studies that apply traditional ML algorithms for brain tumor classification.
- Discuss the features used and their effectiveness.

b. Convolutional Neural Networks (CNNs):

- Explore the application of CNNs in processing medical images for brain tumor detection.
- Review notable CNN architectures and their performance.

Therefore, the focus of our research is directed towards the training and implementation of
lightweight deep learning models. These models are designed to strike a balance between
efficiency and accuracy, making them more suitable for resource-constrained environments, such
as web applications.

3
Research Methodology:
The project will follow the following research methodology:
● Splitting the dataset into test and train.
● Training the model using CNN.
● Building the webapp using Streamlit.

Figure 1: Steps taken to train the model.

Technology Stack:

● AI/ML
● Python
● Sklearn
● OpenCV
● Streamlit

4
References:

1) "Principles of Medical Imaging" by Richard L. Van Metter and Michael L. Williams.

2) "Medical Image Analysis" by Atam P. Dhawan.
3) "Machine Learning: A Probabilistic Perspective" by Kevin P. Murphy.
4) Python Documentation - https://docs.python.org/3/
5) DataSet - https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri
6) OpenCV Documentation - https://docs.opencv.org/4.x/
7) Streamlit Documentation - https://docs.streamlit.io/