Abstract:
Polycystic Ovary Syndrome (PCOS) is an endocrine abnormality that occurred in female reproductive cycle. This paper designed an application to classify Polycystic Ovary S...Show MoreNotes: As originally submitted and published there was an error in this document. The authors subsequently provided the following text: "Author Putria Febriana (School of Computing, Telkom University, Bandung, Indonesia) was omitted from the byline and should be included". The original PDF remains unchanged.
Metadata
Abstract:
Polycystic Ovary Syndrome (PCOS) is an endocrine abnormality that occurred in female reproductive cycle. This paper designed an application to classify Polycystic Ovary Syndrome based on follicle detection using USG images. The first stage of this classification is preprocessing, which employs low pass filter, equalization histogram, binarization, and morphological processes to obtain binary follicle images. The next stage is segmentation with edge detection, labeling, and cropping the follicle images. The following stage is feature extraction using Gabor wavelet. The cropped follicle images are categorized into two groups of texture features: (1) Mean, (2) Mean, Entropy, Kurtosis, Skewness, and Variance. This result in 2 datasets prepared for classification process, i.e. (1) data set A has 40 images that consist of 26 normal images and 14 PCOS-indicated images. It counted by Mean texture feature and obtained 275 follicle images. (2) Dataset B has 40 images consist of 34 normal images and 6 PCOS-indicated images. It counted by Mean, Entropy, Kurtosis, Skewness, and Variance texture features then obtained 339 follicle images. The last stage is classification. It identifies the features of PCO and non-PCO follicles based on the feature vectors resulted from feature extraction. Here, three classification scenarios are designed: (1) Neural Network-Learning Vector Quantization (LVQ) method, (2) KNN — euclidean distance, and (3) Support Vector Machine (SVM) — RBF Kernel. The best accuracy gained from SVM-RBF Kernel on C=40. It shows that dataset A reach 82.55% while dataset B that obtained from KNN-euclidean distance classification on K=5 reach 78.81%.
Notes: As originally submitted and published there was an error in this document. The authors subsequently provided the following text: "Author Putria Febriana (School of Computing, Telkom University, Bandung, Indonesia) was omitted from the byline and should be included". The original PDF remains unchanged.
Published in: 2015 3rd International Conference on Information and Communication Technology (ICoICT)
Date of Conference: 27-29 May 2015
Date Added to IEEE Xplore: 03 September 2015
ISBN Information: