You are on page 1of 12

Identification of Breast cancer

using Deep Learning

• Abstract
• Architecture
• Design Methodologies
• Dataset
• Process Evaluation
• Modern tool utilization
• Adoption of code
• Conclusion
• References

Cancer is the second leading cause of death globally, and was

responsible for 8.8 million deaths in 2015. Globally, nearly Breast
cancer is 25 percent of all the cancers in women. Some
technologies like Machine learning and Deep learning can be
applied in order identify cancer so that necessary treatment can be
given to the patient. Deep learning is preferred to Machine
learning for this analysis for the accuracy in the detection of
cancer. Various attributes are taken from previously calculated
data of digitalized breast mass images. After preprocessing of the
data, Convolutional Neural Networks(CNN) with a minimum of 3
hidden layers can be used for the detection process.
Design Methodology
• The breast cancer dataset is taken from kaggle.

• This dataset contains the extracted features from images

of effected and uneffected breast mass.

• There are 20 attributes and 600 records.

• Some of the attributes are

us_worst,area_worst etc which are computed for each cell
Process Evaluation
• Jupyter Notebook is installed and the required packages such
as numpy, pandas, matplotlib etc are imported.

• Next, the data file was read into the Notebook.

• Required pre-processing techniques are applied to clean the

data and then mapping of “Benign” to “0” and “Malignant” to
“1” will be done.

• Scaling of data is done thus creating high dimensional feature

 A feed forward neural network with required number of
hidden layers will be created.

 After splitting the dataset into training and testing data, then
the model is generated using the training data by performing
cross validation analysis.

 Then the model will be evaluated by using the test dataset.

Modern Tool Utilization
• For implementing this process, we are using python language
in jupyter notebook.
• Jupyter Notebook (Formerly IPython Notebooks) is a web-
based interactive computational environment for creating
Jupyter notebooks documents.
• NumPy is a library for the Python programming language,
adding support for large, multi-dimensional arrays and
matrices, along with a large collection of high-level
mathematical functions to operate on these arrays.
• Pandas is a software library written for the Python
programming language for data manipulation and analysis.
Adoption Of Code

• Till now the dataset is loaded into Jupyter Notebook that

was installed.

• Further, we would like to perform preprocessing

techniques and then data is passed through Feed Forward
Neural Network with required layers.
• Saira Charan, KurramKurshid, “Breast Cancer Detection in
Mammograms using Convolutional Neural

• PengchengXi, ChangSu, “Abnormality Detection in Mammography

using Deep Convolutional Neural Networks”, Conference Paper,2018.

• Alexander Rakhlin , Alexey Shvets, “Deep Convolutional Neural

Networks for Breast Cancer Histology Image Analysis”, ,International
Journal of Applied Engineering Research ISSN,2018.

• S.Karthik, Srinivasa Perumal, “Breast Cancer Classification Using

Deep Neural Networks”, Knowledge Computing and Its