Skip to Main Content

An official website of the United States government

A Machine Learning Approach to Diagnosing Breast Cancer

Principal Investigator
Name
Michele La Ferla
Degrees
B.Sc., B.A.
Institution
University of Malta
Position Title
Masters Student
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCOI-585
Initial CDAS Request Approval
Feb 19, 2020
Title
A Machine Learning Approach to Diagnosing Breast Cancer
Summary
This project will use a machine learning approach to train a model to classify a benign or malign cancer tumor at the earliest stages possible. It will be using the pathological stages obtained from PLCO to train the mentioned model of a convolutional neural network. In particular, this study will take into consideration the patient's family medical history to verify if any of the patient’s family members have a history of breast cancer; and common symptoms with other patients, of which there exist several implementations of solutions from the machine learning field of study.
Aims

Aim

The study will gather data from large datasets containing symptoms of breast cancer, which coupled with the medical history of the patient and their family members, these cancers would be accurately diagnosed at the earliest stage possible.

Objectives

The objectives of this research is to:
- Experiment different machine learning techniques and find out which is the best to predict the diagnosis of breast cancer at an earlier stage.The chosen model will be used for training purposes on the dataset and to build the model.
- Use the family history risk factor to have an insight as to whether a patient is more susceptible to be diagnosed with breast cancer or not.
- Verify the results of the model and its algorithms and extract notable conclusions from this study. These conclusions will then be presented to the Maltese medical community.

Collaborators

Michele La Ferla