An Efficient Classification Of Fetal Ultrasound Lung Based Enhanced Biased Maximum Margin Analysis

Abstract

Objective: Respiratory Distress Syndrome (RDS) is the lung disease which is the main cause of infant death. The identification of fetal Lung disease at initial stages is of extreme importance if it is intended to degrade high mortality rate. It is mandatory to monitor the unborn baby’s functioning of lungs, to check whether the baby can able to breathe on his/her own. Methodology/ investigation: For that reason, there is need to develop a new approach for the analysis of fetal lung based ultrasound images. Image processing seems to be a supportive tool to analyze and classify the lung disease. Initially, the input fetal lung images are forwarded to the preprocessing techniques by Wiener filter, segmentation is performed to segment the fetal lung image by Ostu Thresholding and the features are extracted by Gray-Level Co-Occurrence Matrix (GLCM). The classifier is applied on the extracted features by Enhanced Biased Maximum Margin Analysis (EBMMA), finally the system is classified the fetal lung conditions. Conclusion: In this study, performance results are tested on fetal lung database collected from local hospital and evaluated in terms of accuracy, sensitivity and the Specificity. The obtained results show the efficient accuracy classification for analyse the lung disease when compared to other techniques. This technique can help radiologists and doctors to analyze the maturity condition of fetal lung condition of diseases at early stages and to avoid serious disease.