Skip to Main Content

An official website of the United States government

Government Funding Lapse

Because of a lapse in government funding, the information on this website may not be up to date, transactions submitted via the website may not be processed, and the agency may not be able to respond to inquiries until appropriations are enacted. The NIH Clinical Center (the research hospital of NIH) is open. For more details about its operating status, please visit  cc.nih.gov. Updates regarding government operating status and resumption of normal operations can be found at OPM.gov.

“Prostate cancer data analysis using machine learning methods”

Principal Investigator
Name
Henry Barlow
Degrees
BSc/BA (not yet completed)
Institution
The University of Sydney
Position Title
Undergraduate Student
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCO-468
Initial CDAS Request Approval
Apr 5, 2019
Title
“Prostate cancer data analysis using machine learning methods”
Summary
We aim to build a predictor of high and low grade prostate cancer for multiple datasets. The additional question we seek to answer is “What are the most effective screening methods for prostate cancer?”, by comparing which variables were most successful at predicting cancer across multiple datasets.
Aims

We aim to develop a prediction model for prostate cancer based on variables available in clinical data which may be attained without biopsies, since, as stated in the response of question 365, biopsies are invasive and potentially harmful to patients who do not have prostate cancer. The most common variable used is prostate specific antigen (PSA) levels, however it has been criticised on the grounds that testing with it leads to a lot of false positive diagnoses, meaning it might not actually reduce mortality from prostate cancer, but only increase the amount of people going through unnecessary treatment which has negative side effects such as incontinence [1]. We therefore aim to build a prediction model that will be of use in a clinical setting, and if possible investigate whether variables such as age and free:total PSA can be used to obtain better predictions than total PSA alone, where free PSA is PSA that is not bound to other proteins [2].

[1] Roland Martin, Neal David, Buckley Richard. “What should doctors say to men asking for a PSA test?” BMJ 2018; 362 :k3702
[2] Nancy Ferrari, “What is the difference between PSA and free PSA?,” Harvard Medical School, accessed April 3, 2019, https://www.health.harvard.edu/blog/what-is-the-difference-between-psa-and-free-psa-20091001114.

Collaborators

Dr. Matloob Khushi (Director, Master of Data Science School of Computer Science, J12 - Computer Science Building, The University of Sydney)

Related Publications