Louis, Claudio Michael and Erwin, Alva and Galinium, Maulahikmah (2020) Automated Annotation of in Vitro Fertilization Time-Lapse Using Deep Neural Network. Bachelor thesis, Swiss German University.

Preview	Text Claudio Michael Louis 11602010-TOC.pdf Download (1MB) | Preview
	Text Claudio Michael Louis 11602010-1.pdf Restricted to Registered users only Download (570kB)
	Text Claudio Michael Louis 11602010-2.pdf Restricted to Registered users only Download (7MB)
	Text Claudio Michael Louis 11602010-3.pdf Restricted to Registered users only Download (1MB)
	Text Claudio Michael Louis 11602010-4.pdf Restricted to Registered users only Download (2MB)
	Text Claudio Michael Louis 11602010-5.pdf Restricted to Registered users only Download (232kB)
Preview	Text Claudio Michael Louis 11602010-Ref.pdf Download (632kB) | Preview

Abstract

Infertility is a long living issue that has plagued a significant portion of the global population. To combat this issue, a series of treatment called has been devised over the years, and is collectively referred to as Assisted Reproductive Technology (ART). Among the multiple existing ART, one of the most effective and most commonly utilized in modern times is In Vitro Fertilization (IVF). This however, is only a relative comparison as the actual success rate of this IVF treatment is still low. Despite having gone through a few improvements over the year, a certain step in IVF have remained the same for an extended period, namely the observation and assessment of embryo culture. Embryo culture describes the process of observing embryos during incubation, and assessment is done to chose the best embryo post incubation, where the best ones will be selected for transfer back into the female patient. Recently observation has been improved through time-lapse observation technology, yet assessment is still done manually by an embryologist. In this research is proposed an automated deep learning-based system to automate the process of automation. Using a deep-learning based approach, a system will be created that can automatically annotate and analyze a time-lapse observation video. Utilizing a Convolutional Neural Network approach, it is possible to perform automated annotation on a time-lapse video with an accuracy rate up to 70% for detecting up to the 4-cell stage.

Item Type:	Thesis (Bachelor)
Uncontrolled Keywords:	In Vitro Fertilization; Cell Counting; Deep Learning; Convolutional Neural Network
Subjects:	Q Science > QA Mathematics > QA76 Computer software Q Science > QA Mathematics > QA76 Computer software > QA76.87 Neural networks (Computer science) T Technology > T Technology (General) > T58.5 Information technology
Divisions:	Faculty of Engineering and Information Technology > Department of Information Technology
Depositing User:	Faisal Ifzaldi
Date Deposited:	02 Nov 2020 14:07
Last Modified:	02 Nov 2020 14:07
URI:	http://repository.sgu.ac.id/id/eprint/1864

Actions (login required)

View Item