Skip to Main Content

An official website of the United States government

Principal Investigator
Name
Geetanjali Rave
Degrees
MCA, MSc.(BI), MBA
Institution
Ramaiah Institute of Technology
Position Title
Senior Teaching Assistant
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-890
Initial CDAS Request Approval
Mar 2, 2022
Title
Thyroid carcinoma relation with ovaries using Machine Learning Techniques
Summary
The ovaries generate progesterone and oestrogen and the thyroid gland produces T3 and T4 hormones. The thyroid regulates the human body's metabolism with the use of hormones.
The ovaries and the thyroid gland (the ‘master gland' that regulates metabolism) are inextricably linked. With the thyroid gland, female ovaries contain the essential iodine concentration. Both, the ovaries and the thyroid are affected by iodine deficiency.
Most significantly, the ovary contains thyroid hormone receptors, which play a role in egg formation as well as conception. Excess hormone production results in oestrogen dominance, whilst a lack of hormone results in menstruation and reproductive issues.
This project aims in finding the relation between patients having thyroid cancer and their ovaries (i.e whether their ovaries were removed, if they have had any episode of cancer in the ovaries) using machine learning techniques.
Aims

Aim : Using machine learning techniques understand the relation between thyroid cancer and ovaries based on the below conditional nests :

1) Check Thyroid Cancer occurrence in body in Females only.
2) Then check for the ovarian cancer occurrence :
2a) Incase there has been an episode on ovarian cancer before check details (pertaining to ovary) and establish a connection with the available parameters in the thyroid dataset
2b) Else if there is no episode on ovarian cancer before check details (pertaining to ovary) and establish a connection with the available parameters in the thyroid dataset

3) Based on the available 2a and 2b load filtered data into a Data Frame and perform the spilt of testing and training of data.
4) Get a correlation matrix for the dataset.
5) HyperTune the parameters if necessary
6) Binary classify the dataset (either via , SVM, Random Forest, Decision Tree etc)
7) Learn the clustering of the dataset ( either via K Nearest Neighbour, Random Forest etc)
8) Perform their Regression.

Result : Understand their connection. In the result also check for
a ) precision
b) recall
c) f1-score
d) support
e) accuracy

Collaborators

Geetanjali Rave - Senior Teaching Assistant at the Department of Master of Computer Application, M S Ramaiah Institute of Technology, Bangalore, India.
http://www.msrit.edu/department/faculty-detail.html?dept=mca&ID=14

Dr. S Seema : Professor and Head at the Department of Master of Computer Application, M S Ramaiah Institute of Technology, Bangalore, India.
page link - http://www.msrit.edu/department/mca.html