Detection of Breast Cancer from Histopathology

image and Classifying Benign and Malignant State

Using Machine learning
Content
■ Introduction
■ Current System
■ Block Diagram
■ Methodology
■ Hardware Required
■ Software Required
■ Advantages and Disadvantages
■ Conclusion
■ References
Introduction
■ Breast cancer (BC) is one of the most common cancers among women
worldwide, representing most new cancer cases and cancer-related deaths
according to global statistics, making it a significant public health problem
today.
■ The early diagnosis of BC can improve the prognosis and chance of survival
significantly, as it can promote timely clinical treatment to patients.
■ Further accurate classification of benign tumors can prevent patients
undergoing unnecessary treatments. Thus, the correct diagnosis of BC and
classification of patients into malignant or benign groups is the subject of
much research. Because of its unique advantages in critical features detection
from complex BC datasets, machine learning (ML) is widely recognized as the
methodology of choice in BC pattern classification and forecast modelling.
Challenge
■ Build an algorithm to automatically identify whether a patient
is suffering from breast cancer or not by looking at biopsy
images. The algorithm had to be extremely accurate because
lives of people is at stake.
Breast Tumors Stage
■ Benign tumors
– Aren’t cancerous
– They can often be removed
– In most cases, they do not come back
– Cells in benign tumors do not spread to other parts of the body
■ Malignant tumors
– Are cancerous
– Cells grow out of control
– Cells in these tumors can invade nearby tissues and spread to other
parts of the body
Breast Tumors Stage
Literature Survey
N Papers Authors Year
o
1. On Predicting and Analyzing Breast Masud Rana Basunia;Ismot Ara 2020
Cancer using Data Mining Approach. Pervin;Md. Al Mahmud;Suman
Saha;Mohammad Arifuzzaman
2. Breast Cancer Image Classification on Jiandong Ye;Yihao Luo;Chuang 2019
WSI with Spatial Correlations Zhu;Fang Liu;Yue Zhang
3. Multiclass Breast Cancer Classification Phu T. Nguyen;Tuan T. 2019
Using Convolutional Neural Network Nguyen;Ngoc C. Nguyen;Thuong
T. Le
4. Breast Cancer Detection Based on Deep Nur Syahmi Ismail;Cheab Sovuthy 2019
Learning Technique
5. Breast Cancer detection from Vartika Mishra;Yamini 2019
Thermograms Using Feature Extraction Singh;Santanu Kumar Rath
and Machine Learning Techniques
Architecture
Methodology
■ Typical pipeline of this software consists of loading the images, cropping
images, some image processing modules, identifying the objects and so
on.
■ Firstly, we have cropped the original image and then unmixed that image.
The next step is to convert the unmixed image into color image followed
by a color to gray scale conversion for thresholding and segmentation.
■ After the breast cancer cell detection feature extraction from these images
is done by using a pipeline.
■ Finally, classification has been done with the ANN, SVM and Fuzzy
Logic algorithm in OpenCV and Python.
Algorithms comparison
Algorithm Classification accuracy Computational Time(ms)

SVM 94.8% 82

Adaboost 93.6% 69

CNN 97.93% 23

ResNet 97.92% 32
Dataset
 [Web Link] -
https://web.inf.ufpr.br/vri/databases/breast-cancer-histopatholo
gical-database-breakhis/

Benign sample Malignant sample

Software Required
 Programming language - Python3
 Libraries - OpenCV, NumPy, Pandas, SciPy
 Algorithm – CNN
 Cloud – Amazon AWS
 Web - HTML, CSS, JavaScript, Bootstrap, Flask
Advantages
 Reduces the pathologist’s workload
 No critical region is overlooked
 High Reliability and accuracy
 Low cost of diagnosis
Conclusion
■ Our proposed demonstrates that machine learning models trained in an end-to-
end fashion can be highly accurate and potentially readily transferable across
diverse mammography platforms. Machine learning methods have enormous
potential to further improve the accuracy of breast cancer detection on
screening mammography as the available training datasets and computational
resources expand. Our approach may assist future development of superior
CAD systems that could be used to help prioritize the most suspicious cases to
be read by a radiologist, or as an automatic second reader after making an
initial independent interpretation. Our end-to-end approach can also be
applied to other medical imaging problems where ROI annotations are scarce.
References
1. Breast Cancer detection from Thermograms Using Feature Extraction and Machine Learning Techniques
Vartika Mishra;Yamini Singh;Santanu Kumar Rath 2019 IEEE 5th International Conference for Convergence
in Technology (I2CT) Year: 2019 | Conference Paper | Publisher: IEEE
2. Breast Cancer Detection Based on Deep Learning Technique Nur Syahmi Ismail;Cheab Sovuthy 2019
International UNIMAS STEM 12th Engineering Conference (EnCon) Year: 2019 | Conference Paper |
Publisher: IEEE
3. Breast Cancer Image Classification on WSI with Spatial Correlations Jiandong Ye;Yihao Luo;Chuang
Zhu;Fang Liu;Yue Zhang ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and
Signal Processing (ICASSP) Year: 2019 | Conference Paper | Publisher: IEEE
4. On Predicting and Analyzing Breast Cancer using Data Mining Approach Masud Rana Basunia;Ismot Ara
Pervin;Md. Al Mahmud;Suman Saha;Mohammad Arifuzzaman 2020 IEEE Region 10 Symposium
(TENSYMP) Year: 2020 | Conference Paper | Publisher: IEEE
5. Look, Investigate, and Classify: A Deep Hybrid Attention Method for Breast Cancer Classification Bolei
Xu;Jingxin Liu;Xianxu Hou;Bozhi Liu;Jon Garibaldi;Ian O. Ellis;Andy Green;Linlin Shen;Guoping Qiu 2019
IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019) Year: 2019 | Conference Paper |
Publisher: IEEE
6. https://towardsdatascience.com/building-a-simple-machine-learning-model-on-breast-cancer-data-eca4b3b99fa
3
Thank You !