Breast Cancer Detection.pptx, PPT

1. BREAST CANCER DETECTION
SCHOOL OF COMPUTING SCIENCE AND ENGINEERING
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
GALGOTIAS UNIVERSITY, GREATER NOIDA,INDIA
2021-2022
Submitted By :-
Chandrajeet Jha(19SCSE1010397)
Bipul Raj(19SCSE1010599)
Project ID:BT3148
PROJECT TITLE:BREAST CANCER DETECTION
UnderThe Supervisionof -
Ms. SONIA KUKREJA

2. INTRODUCTION
• Breast cancer signifies unique of the diseases that as more losses each year. Breast cancer is the
utmost collective cancer amongst women universal secretarial for 25% of all cancer cases and
pretentious 2.1 million persons in 2015 primary diagnosis suggestively rises the likelihoods of
persistence.
• The existing methods are Machine learning, method of training machines with data to make the
decision for same conditions and its application can be observed in various domains such as
medical, network, object identification and security etc. There are 2 machine learning types that
is single and hybrid approaches as for instance Support vector machine.
• Breast cancer has been ranked as the most common cancer among Indian women and
accounts
for 27% of all cancers in women. About 1 in 28 women are likely to develop breast cancer
during
their lifetime. In the urban areas, the incidence is 1 in 22 as compared to the rural areas
where 1
in 60 women develop breast cancer.
• In India, the number of breast cancer cases are rising. According to 2018 report of breast
cancer statistics, there are 1,62,468 new registered cases and 87,090 reported deaths.

3. FACTORS:
• Advanced stage of breast cancer diagnosis in India revolves
around two factors:
1.Non-existent of breast cancer awareness programs as
compared to developed countries.
2.Non-participation of women on a mass scale if any such
programs exist.

4. METHODOLOGY:
• The proposed methodology will help us to distinguish between malignant and benign tumor at
a faster rate. CNN being a complex and complicated classifier can extract vital features
automatically without depending on preprocessing. It is more proficient because it filters the
important parameters and also is flexible being capable to work exceptionally well on image
data.
• The importance of feature selection in a machine learning model is inevitable. It turns the data
free from ambiguity and reduces the complexity of the data. Also, it reduces the size of the
easy to train the model and reduces the training time. It avoids over tting of data. Selecting the
feature subset from all the features increases the accuracy. Some feature selection methods
wrapper methods, lter methods, and embedded methods.
• STEPS:
• Select the input image • Pre-processing • Image Augmentation • Building Convolutional
Architecture • Model Testing • Testing for output • Accuracy Testing • Result Comparison with
actual dataset Select the input image

5. Risk factors for breast cancer:
1.Age: The chance of getting breast cancer increases as women age. Nearly 80
percent of breast cancers are found in women over the age of 50
2.History of breast cancer: A woman who had breast cancer in one breast is at an
increased risk of developing cancer in her other breast
3.Family history of breast cancer: A woman has a higher risk of breast cancer if
her mother, sister or daughter had breast cancer, especially at a young age
(before 40). Having other relatives with breast cancer may also raise the risk
4.Genetic factors: Women with certain genetic mutations, including changes to
the BRCA1 and BRCA2 genes, are at higher risk of developing breast cancer
during their lifetime. Other gene changes may raise breast cancer risk as well.
5.Childbearing and menstrual history: The older a woman is when she has her
first child, the greater her risk of breast cancer. Also at higher risk are:
1. Women who menstruate for the first time at an early age (before 12)
2. Women who go through menopause late (after age 55)
3. Women who never had children (nulliparous)

6. SCHEMATIC DIAGRAM

7. CONCLUSION:
• Breast cancer detection using machine learning has achieved successfully with
accuracy up to 97.4%. By using this machine learning the output is effective and
faster and reduces the complexity.
• Here we have used combination of classifiers & algorithms such as decision tree
algorithm, random algorithm and logistic regression helped to achieve high
accurate and efficient model. The results shown in decision tree classifiers
prediction and in the actual classification of the patients which presenting ones
as malignant (cancerous) and zeros as benign (non- cancerous).
• This model can predict a greater number of correct values than negatives. By
detecting the breast cancer at early stage, the cancer can be curable and the
patients can avoid painful surgeries. The overall computational time for the
preprocessing would be 3.5sec & the time for the processing stage would be
around 5sec for the number of dataset considered, this time could vary
depending upon the number of dataset that has been chosen.

8. RESULT:
• The analysis of traditional ML reveals the limited usage of the
methods, whereas the DL methods have great potential for
implementation in clinical analysis and improve the diagnostic
capability of existing CAD systems.
•