Study and detection of PCOS related diseases using CNN

Polycystic ovary syndrome (PCOS) is a group of symptoms caused by high levels of androgens in women. The cause of PCOS is a group of genetic and environmental factors that are common pathologies, often associated with clinical symptoms of arteries, hirsutism, acne, and hyperandrogenism, along with c...

Full description

Saved in:
Bibliographic Details
Published in:IOP conference series. Materials Science and Engineering 2021-02, Vol.1070 (1), p.12062
Main Authors: Sumathi, M, Chitra, P, Sakthi Prabha, R, Srilatha, K
Format: Article
Language:English
Subjects:
Algorithms
Arteries
CNN
Cysts
Datasets
Deep learning
Feature extraction
Image classification
Image processing
Infertility
Medical image processing
Medical imaging
Ovaries
PCOS related diseases
Polycystic ovary syndrome
Programming languages
Signs and symptoms
Ultrasonic imaging
Ultrasound
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Polycystic ovary syndrome (PCOS) is a group of symptoms caused by high levels of androgens in women. The cause of PCOS is a group of genetic and environmental factors that are common pathologies, often associated with clinical symptoms of arteries, hirsutism, acne, and hyperandrogenism, along with chronic infertility. Recent studies show that about 18% of Indian women suffer from this syndrome. Doctors were manually examining ultrasound images and conclude the affected ovary but unable to find whether it is a simple cyst, PCOS, or cancer cyst. In this paper, CNN based algorithms proposed and coding developed in Python programming for classification of cysts, and they are filled with blood or fluid using ultrasound images. The study is performed on CNN based image processing feature extraction to classify cysts in the dataset. That is the study is carried out using an independent trained dataset of the same PCOS related diseases. Finally, the test dataset is used for performing the feature extraction process and the results are met with 85% accuracy using performance factors.
ISSN:1757-8981
1757-899X
1757-899X
DOI:10.1088/1757-899X/1070/1/012062