Prediction of polycystic ovary syndrome with statistical models

Qianhui Ma

doi:10.1117/12.2668992

24 March 2023 Prediction of polycystic ovary syndrome with statistical models

Qianhui Ma

Proceedings Volume 12611, Second International Conference on Biological Engineering and Medical Science (ICBioMed 2022); 126114T (2023) https://doi.org/10.1117/12.2668992
Event: International Conference on Biological Engineering and Medical Science (ICBioMed2022), 2022, Oxford, United Kingdom

Abstract

Polycystic ovary syndrome (PCOS) is a common hormone disorder in women. Women with PCOS usually suffer from irregular menstruation, obesity, and excessive hairiness. The number of diagnoses has been increasing in recent years, but there are deficiencies in the diagnosis of PCOS. In this paper, the study was conducted to investigate the most relevant factors for the diagnosis of PCOS and to develop statistical models for the prediction of PCOS. The methodology of the research includes obtaining the dataset, analyzing the data, making a prediction model by Logistic Regression and K-Nearest Neighbors Algorithm (KNN), and calculating the accurate rate of each model. By analyzing the data, it can be found that follicle number, skin darkening, hair growth, weight gain, and cycle irregularity are important to determine whether the female has PCOS. The accuracy of the Logistic Regression was 89.72%, and the accuracy of the KNN was 93.46%. Both models did well in prediction, but the KNN model has a better performance and higher prediction accuracy than the Logistic Regression. The above predictors are the most common factors among patients with PCOS. Therefore, when confirming a diagnosis of PCOS, doctors may focus primarily on these factor.

Citation Download Citation

Qianhui Ma "Prediction of polycystic ovary syndrome with statistical models", Proc. SPIE 12611, Second International Conference on Biological Engineering and Medical Science (ICBioMed 2022), 126114T (24 March 2023); https://doi.org/10.1117/12.2668992

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available