Presentation is loading. Please wait.

Presentation is loading. Please wait.

Submitted By : Pratish Singh Kuldeep Choudhary Chinmay Panchal

Similar presentations


Presentation on theme: "Submitted By : Pratish Singh Kuldeep Choudhary Chinmay Panchal"— Presentation transcript:

1 Statistical modeling of breast cancer in the light of data mining and artificial intelligence
Submitted By : Pratish Singh Kuldeep Choudhary Chinmay Panchal Kirtesh Jain Vickey Kapoor Faculty Guide : Dr. Prasun Chakrabarti

2 Features and Prediction of Breast Cancer – A Review
Breast cancer is a malignant tumor that develops when cells in the breast tissue divide and grow without the normal controls on cell death and cell division . Breast cancer is the most common major cancer in women and the fifth major cause of cancer deaths in women. Among the most commonly used research methods for breast cancer are laboratory studies, observational studies and clinical trials.

3 Knowledge Discovery in databases and Data Mining
Advancements in technologies have great impact on data collection methodologies. KDD encompasses variety of statistical analysis, pattern recognition and machine learning techniques. Applications of data mining have been used to provide benefits to many areas of medicine, including diagnosis, prognosis and treatment.

4 Types of breast cancer Ductal carcinoma Invasive Ductal carcinoma
Invasive Lobular carcinoma Inflammatory breast cancer

5 Treatment methods Lumpectomy (breast conserving)
Mastectomy (breast removal) Chemotherapy Radiation therapy Hormonal therapy

6 Mammograms A mammogram, which uses a series of X-ray images of your breast tissue, is currently the best imaging technique for detecting tumors.

7 Agglomerative Clustering
APPLICATION OF MATHEMATICAL MODELLING AND ANALYSIS IN THIS PERSPECTIVE - A R&D APPROACH K-Means Clustering Agglomerative Clustering Mapping scheme on the basis of projected window Decision Trees Optimum cell therapy selections on the basis of boundary scan Detection of trend of cancer spread on the basis of curve fitting and its related error estimation

8 K-means clustering A non-hierarchical approach to forming good clusters is to specify a desired number of clusters, say, k, then assign each case (object) to one of k clusters so as to minimize a measure of dispersion within the clusters. A very common measure is the sum of distances .

9 Proposed concept based on clustering
The affected cells are extracted based on boundary scan and then graphically plotted. Based on curve fitting approach spread of path of cancer can be diagnosed. The cells in each cluster will be examined such that whether they act as parent node of the other in such context decision trees are used.

10 Agglomerative Clustering
bottom-up clustering method The hierarchy within the final cluster has the following properties: Clusters generated in early stages are nested in those generated in later stages. Clusters with different sizes in the tree can be valuable for discovery.

11 Mapping scheme on the basis of projected window
affected and non-affected cells are clearly identified and the effect of distortion of cell is quantified. We can apply the K-means clustering concept for the grouping specially when there is significant variety of distortion of cell. Two parameters representing distortion percentage are to be taken into consideration and the clusters are formed on the basis of their respective probability density function. In this context if there is a newly affected cell at a specific timing instant, in that case on the basis of agglomerative clustering that particular cell will be considered as an element of that particular cluster

12 Decision trees model of computation or communication
a sequence of branching operations based on comparisons of some quantities

13 Use of decision tree For predicting the cancer possibility.
For predicting the type of cancer. For predicting the extension of cancer affect.

14 Optimum cell therapy selections on the basis of boundary scan
Cell with greater density of adjacent affected cells, will be diffused

15 Detection of trend of cancer spread on the basis of curve fitting and its related error estimation
Analysis of cancer at different intervals Curve prediction

16 Fig. : Affected cells at time interval t3
Fig. : Affected cells at time interval t Fig. : Affected cells at time interval t2 Fig. : Affected cells at time interval t3 Fig. : Graph showing number of cells affected at different timing intervals during various phases of cancer

17 Conclusion Detection of cancer cells through boundary scanning
Analysis of cancer spread on curve fitting and data mining concepts Proposing the equation for applying chemotherapy

18 References http://www.breastcancer.org/ http://www.facd.info/
full html

19 Thank you !!


Download ppt "Submitted By : Pratish Singh Kuldeep Choudhary Chinmay Panchal"

Similar presentations


Ads by Google