Scribd
Upload
11 views17 pages

Presentation 3

ppt on cancer

Uploaded by

shailzakanwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views17 pages

Presentation 3

ppt on cancer

Uploaded by

shailzakanwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Application of Machine

Learning in breast cancer


Content
o Introduction about Machine learning
o Introduction about cancer
o Cancer prediction using machine learning
o Process steps in ML
o Literature review
o Summary of ML algorithm used in cancer
o Future of Cancer Prognosis Look Like
o Reference
Introduction about Machine
learning
ML is said to learn from experience E, w.r.t to some class of task T and a performance
measure P, if learners performance at the task in the class as measured by P improves with
experiences.

Notation of data set-


◦ labeled dataset-

◦ Unlabeled dataset
Categorization of machine
learning-
•ML is categorized in four part- Classification
• Supervised learning Supervised
learning
• Used labeled data set.
Regression
• Find the relationships between the feature set and the label set.
• Unsupervised learning
• Used unlabeled data set. Clustering
• Output is not given Unsupervised
learning
• Reinforcement learning
Association
• Used to solve the problem of decision making.

• Semi-supervised learning
• Used labeled and unlabeled data.
Introduction about cancer
Cancer is derived from the word ”CRAB”
The term cancer specifically refers to a new growth which has the ability to invade surrounding
tissues, metastasize (spread to other organs).

Tumor is simply refers to a mass. A cancer is a particularly threatening type of tumor.

Fig: Cancer
Types of Tumors
Benign
Benign tumors are non-
malignant/non-cancerous tumor

Tumor Malignant
Malignant tumors are cancerous
growths.
A cancer is another word for a
malignant tumor
Question

How can we use Machine Learning in cancer


Prediction?
Cancer prediction using machine learning
New
Training pattern

Test, Prediction
Training Machine
Patterns + Learning Classifier
class labels Algorithm
Feature
selection Model Predicted
selection Class

fig. – Building Classifier- a flow chart


Process steps in machine
learning

Performance
Feature Training/Testing
Data set Pre-processing evaluation Deployment
extraction model
matrices
Performance Measure Definition
TP (True Positive) Correctly classified Benign
tuples.
1. Confusion matrix
2. Sensitivity=TP/(TP+FN)
FP (False Positive) False classified Malignant
tuples that are marked Benign. 3. Specificity=TN/(FP+TN).
4. Accuracy=(TP+TN)/(TP+FP+FN+TN)
TN (True Negative) Correctly classified Malignant
tuples.

FN (false negative) False classified Benign tuples


that are marked Malignant.
Literature Review:-
1. Kashif, Muhammad, et al
[1] used the MIAS dataset
Author
Mammograms:-
• X-ray images of the breast which
have abnormalities.
• This process can reduce deaths.
• X-ray beam is applied to the breast
which compressed between two
plates to take the image.

Pre-processing:-
• To enhance the image
• Noise are removed from a mammogram.
Segmentation :-
• To find the region of interest
Fig- Breast cancer prediction system[1]
2. Rana et al. [3]
Author used UCI depository (WDBC data set)

No. ML algorithm Methodology


1 Support vector machine • SVM used for predictive analysis
• When there is separable data set then
prefer linear kernel otherwise used
Gaussian.
2. Logistic Regression • It used sigmoid function to carry out
the classification.
• Binary logistic regression can have two
possible outcomes “0” and “1”.
3. KNN • It can used Euclidean or Manhattan
distance.
3. Suji and Rajagopalan, et al [2]
•Author used WDBC data set
•It is the hybrid system
•Hybrid system is the combination of CFRSFS, K-mean clustering and SVM
•CFRSFS and K-mean is used for enhancing the feature and SVM is used for classifying the feature
selected breast cancer dataset.
N Author Technique/algorithm Dataset Accuracy
o.
1. Kashif, SVM MIAS 90
Muhammad, et Logistic Regression 85
al KNN 76
2. Suji and K-Mean, CFRSFS Wisconsin Diagnostic Breast 99.54
Rajagopalan, et Cancer (WDBC)
al
3. Rana et al. KNN WDBC 72
SVM 68
Logic Reg. 68
ML Algorithm used in cancer[6]
Machine learning algorithm Benefits Assumption/or limitation
Decision tree • Easy to understand and efficient • Classes must be mutually exclusive.
(Quinlan, et al) algorithm • Final decision tree depend upon the
• Order of training instance has no effect order of attribute selection.
on training. • Error in training set can result in overly
complex decision tree.
Naïve Bayes • Foundation is based on statically • Assume attribute are statically
(Langley, et al) modelling. independent.
• Easy to understand and efficient • Redundant attribute must mislead the
algorithm classification.
• Useful across multiple domain • Attribute and class frequencies affect
accuracy
Neural Network •Can be used for classification and •Difficult to understand structure of
(Rummelhart et al ) regression. algorithm.
•Able to represent Boolean function (AND, •Two many attributes can result in
OR, NOT) overfitting.
•Tolerance of noise
•Instance can be classified by more than
one input.
Future of Cancer Prognosis
Look Like
You perform clinical tests, either at a clinic or at home. Data is inputted into a pathological ML
system. A few minutes later, you receive an email with a detailed report that has an accurate
prediction about the development of your cancer.
While you might not see AI doing the job of a pathologist today, you can expect ML to replace
your local pathologist in the coming decades, and it’s pretty exciting!
References
1. Kashif, Muhammad, et al. "Application of machine learning and image processing for detection of breast cancer." Innovation in
Health Informatics. Academic Press, 2019. 145-162.
2. Suji, R. Jaya, and S. P. Rajagopalan. "A novel hybrid system for diagnosing breast cancer using fuzzy rough set and LS-
SVM." Research Journal of Applied Sciences, Engineering and Technology 10.1 (2015): 49-55.
3. Rana, M., Chandorkar, P., Dsouza, A., & Kazi, N. (2015). Breast cancer diagnosis and recurrence prediction using machine
learning techniques. IJRET: International Journal of Research in Engineering and Technology eISSN, 2319-1163.
4. Mihaylov, Iliyan, Maria Nisheva, and Dimitar Vassilev. "Application of Machine Learning Models for Survival Prognosis in Breast
Cancer Studies." Information science 10.3 (2019): 93.
5. Aha, David W. "Tolerating noisy, irrelevant and novel attributes in instance-based learning algorithms." International Journal of
Man-Machine Studies 36.2 (1992): 267-287.
6. Cruz, Joseph A., and David S. Wishart. "Applications of machine learning in cancer prediction and prognosis." Cancer
informatics 2 (2006):112.
7. Patrick, Edward A., and Frederick P. Fischer III. "A generalized k-nearest neighbor rule." Information and control 16.2 (1970): 128-
152.
8. Abreu, P.H.; Santos, M.S.; Abreu, M.H.; Andrade, B.; Silva, D.C. Predicting Breast Cancer Recurrence Using Machine Learning
Techniques: A Systematic Review. ACM Comput. Surv. 2016, 49, 52.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy