Fundamental Study On Proposed Cad Based Paediatric Hydronephrosis Classification Model Framework By Enhanced Seed Pixel Region Growing Segmentation And Logistic Regression Classifier Algorithm

.

Authors

  • N.Venkatesa Mohan Senior Assistant Professor, Department of Pediatric Surgery, Coimbatore Medical College and Hospital, Coimbatore, Tamilnadu
  • Valliappan Raman Lecturer, FECS, Swinburne University of Technology Sarawak, Kuching, Sarawak, Malaysia
August 8, 2016

Downloads

Paediatric Hydronerphrosis (HN) is a disease occurs in urinary tract by dilation of kidney (i.e., “water inside the kidney and swollen”) in children’s. Hydronephrosis can be caused by conditions that the child is born with congenital or by conditions that develop after birth. The main challenge of 2D Ultrasound image of hydronephrosis dataset have more noise, and hard to predict the renal dilation in the image through manual process during early stage. In literature, a large number of computer aided diagnostic (CAD) systems using different image modalities, such as ultrasound (US), magnetic resonance imaging (MRI), computed tomography (CT), and radionuclide imaging, have been proposed for early detection of kidney diseases but not for hydrpnephrosis. The main objective of the study is to design new algorithm and develop a computer aided diagnosis framework for early detection of hydronephrosis in children’s to avoid the risk level. The proposed work in this paper is to design a complete framework and new algorithm for segmentation and classification of renal structures which consist of image acquisition, image enhancement, segmentation, feature extraction and classification, whereas in initial stage, ultrasound of kidney image is diagnosed and identify the distortion of renal parenchyma structure caused by the obstruction of urinary tract and abnormality are studied. From the automatic segmentation (i.e. proposed seed pixel region growing method) of renal parenchyma and collecting system in 2D ultrasound images, an optimal set of morphological descriptors (i.e feature extraction) of the kidney are automatically extracted and used as inputs features of a machine learning algorithm. The classifier (i.e proposed logic regression classifier) is able to predict the degree of hydronephrosis of the renal unit, identifying with maximum sensitivity the severe cases that require immediate medical attention, while maximizing the number of noncritical cases where diuretic renography can be avoided. The enhanced seed pixel region growing segmentation and logic regression classification helps to diagnose the presence of hydronephrosis (HN), which leads to an early detection of swollen kidney in children’s and improve the accuracy of risk predication rate.