Medical Image Processing: Detection and Prediction of PCOS – A Systematic Literature Review

Abstract

Purpose: Considered as the most common hormonal disorder among women, polycystic ovary syndrome or PCOS affects 1 in 10 reproductive aged women (18 - 44 years). Ultrasonography is applied for assessing the ovaries to detect PCOS. The patients affected by PCOS consist of 10-12 cysts present in the ovary, but more than 10 cysts are more enough to diagnose the disorder from the ultrasound images. Then, by examining the ultrasound the presence of follicles will be determined. Therefore, the image processing approaches have assisted to identify the characteristics like follicle size, number of follicles and structure to minimize the workload and time of doctors. PCOS do not have better treatment and effective diagnosis. This paper includes reviewing a summary of some of the researches that have been going in area of medical diagnosis. Based on the review, research gap, research agendas to carry out further research are identified.

Approach: A detailed study on the algorithms used in medical image processing and classification.

Findings: The study indicated that most of the classification of polycystic ovarian syndrome is done merely on the clinical data sets. The new hybrid methodology proposed will be more precise as both images and lifestyle are analysed.

Originality: The type of data required for detection system are studied and the architecture and schematic diagram of a proposed system are included.

Paper Type: Literature Review.