Friday 26th of April 2024
 

Artificial Neural Network for Transfer Function Placental Development: DCT and DWT Approach


Mohammad Ayache, Mohamad Khalil and Francois Tranquart

The aim of our study is to propose an approach for transfer function placental development using ultrasound images. This approach is based to the selection of tissues, feature extraction by discrete cosine transform DCT, discrete wavelet transform DWT and classification of different grades of placenta by artificial neural network and especially the multi layer perceptron MLP. The proposed approach is tested for ultrasound images of placenta, resulting in 75% success rate of classification using DCT and 92% using DWT. The method based on multi resolution decomposition analysis and on supervised neural network technique MLP, seems a good method to study the transfer function of placental development in ultrasound.

Keywords: Placenta, discrete cosine transform, discrete wavelet transform, neural network, MLP.

Download Full-Text


ABOUT THE AUTHORS

Mohammad Ayache
Mohammad Ayache obtained a bachelor of engineering in biomedical from the Islamic University of Lebanon. He received the DEA in Signals and Images in biology and medicine from the University of Angers, France in 2004. He received the Ph.D degree in medical Image Processing from the University of Tours, France, in 2007. He is the coordinator of the department of biomedical at the faculty of engineering at the Islamic University of Lebanon. His research interests include advanced neural networks software development and advanced signal and image processing techniques.

Mohamad Khalil
Mohamad Khalil obtained an engineering degree in electrical and electricity from the Lebanese University, faculty of engineering, Tripoli, Lebanon in 1995. He received the DEA in biomedical engineering from the University of Technology of Compiegne (UTC) in France in 1996. He received his Ph.D from the University of Technology of Troyes and University of Technology of Compiegne in France in 1999. He received his HDR (Habilitation à diriger des recherches) from UTC in 2006. He is currently Professor at the Lebanese university, Faculty of engineering, section 1 Tripoli. He is researcher in the Lebanese University and his current interests are the signal and image processing problems: detection, classification, analysis, representation and modeling of non stationary signals, with application to biomedical signals and images. He is Responsible of the research axis (Image and Data Transmission) in LaMA Laboratory, Faculty of sciences, Lebanese university. Director of the center Azm of research in biotechnology and applications in the Lebanese university.

Francois Tranquart
François Tranquart (MD, PhD) was born in Chatillon sur Indre (France) in 1958. He is General Manager of Bracco Suisse SA, Geneva Research Center and Manufacturing site since 2010. This site is dedicated to the development of new ultrasound contrast agents as well as developing solutions to support adequate in vitro and in vivo use of those agents. As a full professor of biophysics, he worked previously at Tours University Hospital as head of Ultrasound Medical Dept (with 100 exams performed daily), head of Centre For Innovation (new developments) and head of a research team (INSERM U930) dedicated to the development of new ultrasound approaches including microbubbles. His main interest is for new diagnostic agents or applications as well as new therapeutic developments involving the use of bubbles. He has presented more than 400 papers in national and international conferences, written more than 120 papers and participated to 20 books.


IJCSI Published Papers Indexed By:

 

 

 

 
+++
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »