DETECTION A PLEURAL EFFUSION OF THORACIC WITH NEURAL NETWORK BACKPROPAGATION METHOD BY FEATURE EXTRACTION BINARY

Situmorang, Elvira and Adi, Kusworo and Setiawati, Evi DETECTION A PLEURAL EFFUSION OF THORACIC WITH NEURAL NETWORK BACKPROPAGATION METHOD BY FEATURE EXTRACTION BINARY. Proceeding The 4th International Seminar on New Paradigm and Innovation of Natural Sciences and its Application (ISNPINSA) . pp. 70-76.

[img]
Preview
PDF - Published Version
6Mb

Official URL: http://isnpinsa.fsm.undip.ac.id/

Abstract

In normal condition the pleural cavity containing 5 ml to 20 ml of fluid. This liquid will be used as lubricants in the process of respiration take place.If the increased accumulation of fluid in the pleural cavity , indicating the presence of a disease called pleural effusion.Pleural effusionincreased accumulation will push the heart position , making it difficult patients to breathe. Difficult of distinguish the excess fluid in the pleural cavity and normal fluid boundary in the pleural cavity is to be minimized by a doctor of radiology. This study is important because it aims (due to) to detect lung pleural effusion with Artificial Neural network method (ANN) back propagation through the binary feature extraction that reduce future physicians in the treatment of patient’s doubts.. Binary feature extraction is obtained from the level set segmentation. The process of image enhancement by histogram equalization and contrast enhancement must be done before the contrast level set segmentation process.. Characteristic binary pattern trained of ANN tissue taken from 5% to 25% in sinus costophrenicus angle image of the thorax.Neural network can recognize the characteristic patterns of the binary feature 15% are well trained. Validation ANN pattern recognition by up to 100%, while process of testing the ANN is able to identify 14 data from 15 test data to test validation value reaches 93.33% on the condition of setting 2 hidden layers, each of hidden layer contain 10 neurons. Keywords: Pleuraleffusion, extraction of Binaryfeature, ANN, Histogram segmentation level set.

Item Type:Article
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering
Q Science > QC Physics
Divisions:Faculty of Science and Mathematics > Department of Physics
ID Code:67737
Deposited By:Mr Instrumentasi MIPA
Deposited On:13 Dec 2018 07:14
Last Modified:13 Dec 2018 07:14

Repository Staff Only: item control page