Scribd
Upload
115 views10 pages

IICS

This document summarizes a presentation on using deep learning approaches for breast cancer detection. It discusses how digital pathology and techniques like convolutional neural networks, recurrent neural networks, and deep belief networks are used in cancer studies. Classification algorithms like decision trees and random forests were applied to a breast cancer dataset containing 569 records and 32 features. The results show the decision tree model achieved 100% training accuracy while the random forest model achieved 99.49% training accuracy.

Uploaded by

prashantrinku
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
115 views10 pages

IICS

This document summarizes a presentation on using deep learning approaches for breast cancer detection. It discusses how digital pathology and techniques like convolutional neural networks, recurrent neural networks, and deep belief networks are used in cancer studies. Classification algorithms like decision trees and random forests were applied to a breast cancer dataset containing 569 records and 32 features. The results show the decision tree model achieved 100% training accuracy while the random forest model achieved 99.49% training accuracy.

Uploaded by

prashantrinku
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/ 10

INTERNATIONAL CONFERENCE

on
INTELLIGENT COMPUTING AND SECURITY: A PARADIGM SHIFT
(IICS2021)

Organized by:
Department of
Master of Computer Applications
G.L. Bajaj Institute of Technology & Management, Greater Noida

Deep Learning Approach for Breast Cancer Detection

Presented by:
Prashant Ahlawat

Authors: Prashant Ahlawat, Mukul Gupta


Manoj Kumar Sharma, Hitesh Kumar Sharma
Digital Pathology
• Newly computer based methods are evolved in Digital Pathology.
• In cancer study these techniques has played a vital role in reducing
human interference and save the results for future cancer diagnosis.
• Convolution neural network, Recurrent neural network, deep belief
network etc. are common techniques which are used in data science.

Structure of cell nucleus


Algorithms

Classification of Benign and Malignant


patient from given dataset

Decision Tree Algorithm of Machine


Learning

Random Forest Algorithms of machine


learning
Dataset Description

5.smoothness_mea 6.compactness_mea
1.radius_mean 2.texture_mean 3.perimeter_mean 4.area_mean
n n

1.count 569.000001 569.000001 569.000001 569.000001 569.000001 569.000001

2.mean 14.12729 19.28964 91.96903 654.88910 0.09636 0.104340

3.std 3.524040 4.301030 24.298980 351.914120 0.014060 0.052810

4.min 6.981001 9.710001 43.790001 143.500001 0.052631 0.019381

5.25.5% 11.700001 16.170001 75.170001 420.300001 0.086371 0.064921

6.50.5% 13.370001 18.840001 86.240001 551.100001 0.095871 0.092631

7.75.5% 15.780001 21.800001 104.100001 782.700001 0.105301 0.130401

maximum 28.110001 39.280001 188.500001 2501.000001 0.163401 0.345401

For this study the dataset used is from Breast Cancer Wisconsin (Diagnostic) Data Set.
This dataset contain total 569 records. It contains 32 columns which represents 32
features of cancerous/non-cancerous cell image. Fine Needle Aspirate (FNA) of a
breast mass is used for computing image features. Given 569 image records contain
357 benign and 212 malignant classification of breast tissue image.
Correlation map
Heat Map for correlation between
parameters
Results and Comparison

• [1]Decision Tree Training Accuracy: 1.0


• [2]Random Forest Training Accuracy: 0.9949748743718593
MODEL [1] :- Decision Tree
AP(1) AN (0)

PP (1) 100 5

PN (0) 7 59

MODEL [2] :- Random Forest


  Actual Positive (1) Actual Negative (0)

PP (1) 103 2

PN (0) 7 59

• PP= Predicted Positive


• PN= Predicted Negative
Accuracy and Loss

Fig. Training and Validation Accuracy and Loss


Thank You

You might also like