Preliminary Analysis of Ultrasound Features for Detection of Polycystic Ovary Syndrome (PCOS) in Women

Polycystic Ovary Syndrome (PCOS) is a prevalent endocrine condition affecting reproductive-aged women thus early diagnosis of PCOS is crucial for effective planning on the treatment. Conventional diagnostic approaches frequently rely on subjective interpretation, which can lead to discrepancy and incorrect diagnosis. Modern computerized systems and tools can be an advantage, thus analysis towards the images is needed to better understand on the features of the images. This study performs the preliminary analysis on ultrasound PCOS and NONPCOS images. After pre-processing using different filters, extraction of the features was performed using intensity histogram, Gray-Level Co-occurrence Matrix (GLCM), and Gray-Level Run Length Matrix (GLRLM) features on PCOS and NONPCOS images. The findings show that combining various feature extraction approaches creates a strong framework for analysing ultrasound PCOS images, improving diagnosis accuracy for PCOS. The findings highlight on which features that could be further analysed, opening the way for better clinical decision-making and patient outcomes in PCOS diagnosis and therapy.