PREDICTION OF TIME-TO-EVENT OUTCOMES IN DIAGNOSING LUNG CANCER

BASED ON SVM AND COMPARE THE ACCURACY OF PREDICTED CNN MODEL.

-USING MACHINE LEARNING

SRINIVASA INSTITUTE OF ENGINEERING AND TECHNOLOGY

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

Under the guidance of

Mrs. V Sai Priya
Submitted by:
DONGA CHANDINI (216N1A0577)
CHEVALA PRASANTHI (216N1A0572)
DWARAMPUDI K S AVINASH REDDY (216N1A0581)
KOLLURI NAGA KARTHIKEYA (216N1A0596)
INTRODUCTION

 Lung cancer, primarily affecting smokers (85% of cases), is one of the leading causes of death globally,
including in Malaysia. Early detection is crucial to improve survival rates, as malignant lung nodules can
grow rapidly and cause fatalities. CT scans are preferred for diagnosis due to reduced noise. The best
technique for feature extraction, object classification, and medical imaging has been demonstrated to be
deep learning. Despite advancements, traditional CNN models often underperform with small medical
datasets, posing challenges in achieving accuracy. This study proposes a hybrid CNN-SVM architecture,
designed to enhance precision by eliminating irrelevant information during the classification process
ABSTRACT

 The leading cause of cancer-related death is lung cancer. The windpipe, main airway, or lungs can all be
the site of lung cancer onset. It results from the unchecked growth and spread of some lung cell types.
Lung cancer diagnosis rates are higher in people with lung conditions like emphysema and a history of
chest pain. The main risk factor for developing lung cancer in Indian men is excessive tobacco use, which
includes smoking cigarettes and beedis. However, Indian women do not smoke as frequently, suggesting
that there may be additional risk factors. Exposure to radon gas, air pollution, and chemicals at work are
additional risk factors. Early lung cancer detection has become crucial and simple thanks to image
processing and deep learning techniques. Images from lung patient Computer Tomography (CT) scans are
used in this study to identify and categorise lung nodules as well as to determine their level of malignancy.
In this project, the CNN algorithm is being used to identify lung cancer in CT-SCAN images, and the
dataset of CT-SCAN images was used to train the CNN algorithm.
SYSTEM ANALYSIS

EXISTING SYSTEM

 Detection of lung cancer in its early stages could reduce the fatality rate. Early detection of cancerous cells is crucial
to prevent them from growing and remove them early before they start to grow rapidly. Also, accurate location and
size detection of benign cells is another crucial point to be taken as in CT scans. It is difficult to differentiate
cancerous nodules from other pulmonary nodules that are not cancerous by only depending on visual estimation,
which is a traditional way for healthcare workers to examine CT scans. Computer-Aided Detection (CAD) systems
have been utilized to process medical images such as CT scans images to assist the medical personnel in the early
detection of lung cancer. CAD systems are based on computer algorithms that process images such as CT scan
images of the lung to determine whether a given image has any sign of cancer; this process is called image
classification.
DISADVANTAGES

 Several machine learning algorithms have been proposed to classify medical images and determine the
existence of lung cancer. Singh & Gupta [19] have researched various machine learning approaches for the
classification and detection of lung cancers using CT scan images. Different classifiers are put to the test to
determine the best machine learning algorithm from the classification process. The results show that
multilayer perception or neural networks can be applied to classify lung cancer CT scan images because it
obtained a high percentage of accuracy precision value. Over the years, deep learning has been used
increasingly in various fields, and researches on deep learning are growing rapidly. Various techniques are
proven to give accurate and precise results, which help humans, detect errors and mistakes at early stages and
help to predict the outcomes.
PROPOSED SYSTEM:

 Lung cancer is one of the leading causes of death worldwide. Early detection of this disease increases the
chances of survival. Computer-Aided Detection (CAD) has been used to process CT images of the lung to
determine whether an image has traces of cancer. This paper presents an image classification method based
on the hybrid Convolutional Neural Network (CNN) algorithm and Support Vector Machine (SVM). This
algorithm is capable of automatically classifying and analyzing each lung image to check if there is any
presence of cancer cells or not. CNN is easier to train and has fewer parameters compared to a fully
connected network with the same number of hidden units. Moreover, SVM has been utilized to eliminate
useless information that affects accuracy negatively. In recent years, Convolutional Neural Networks (CNNs)
have achieved excellent performance in many computer visions tasks. In this study, the performance of this
algorithm is evaluated, and the results indicated that our proposed CNN-SVM algorithm has been succeed in
classifying lung images with 97.91% accuracy. This has shown the method's merit and its ability to classify
lung cancer in CT images accurately.
ADVANTAGES

 Early lung cancer detection has become crucial and simple thanks to image processing and deep learning
techniques. Images from lung patient Computer Tomography (CT) scans are used in this study to identify and
categorize lung nodules as well as to determine their level of malignancy. In this project, the CNN algorithm
is being used to identify lung cancer in CT-SCAN images, and the dataset of CT-SCAN images was used to
train the CNN algorithm. The primary goal of this study is to investigate the performance of a classification
algorithm to aid in the early diagnosis of lung cancer.
 SYSTEM SPECIFICATION:

 HARDWARE REQUIREMENTS:

 System : Intel i3 to untill

 Hard Disk : 100 GB minimum.

 Monitor : 14/10/12/15’ Colour Monitor.

 Mouse : Optical Mouse.

 Ram : 4GB MINIMUM.

 SOFTWARE REQUIREMENTS:

 Operating system : Windows 7/8/10/11.

 Coding Language : Python 3.7.

 Type of Application : GUI Application

 Front-End Technologies : Tkinter API

 Backend Technologies : Tensorflow,Keras,matplotlib,pandas,opencv-pthon,sckit-learn..etc

 IDE Tool : PyCharm community edition 2021

 Dataset : Massey University Dataset (MUD)

THANK YOU