PENERAPAN DEEP LEARNING PADA IMPUTASI DATA TANDA VITAL UNTUK MENINGKATKAN AKURASI PREDIKSI HENTI JANTUNG PADA PASIEN UNIT PERAWATAN INTENSIF

JIORNMIA, THESA and Firdaus, Firdaus (2022) PENERAPAN DEEP LEARNING PADA IMPUTASI DATA TANDA VITAL UNTUK MENINGKATKAN AKURASI PREDIKSI HENTI JANTUNG PADA PASIEN UNIT PERAWATAN INTENSIF. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011281823031.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011281823031_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
Preview	Text RAMA_56201_09011281823031_0221017801_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_56201_09011281823031_0221017801_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (354kB) | Request a copy
	Text RAMA_56201_09011281823031_0221017801_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (799kB) | Request a copy
	Text RAMA_56201_09011281823031_0221017801_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (709kB) | Request a copy
	Text RAMA_56201_09011281823031_0221017801_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (186kB) | Request a copy
	Text RAMA_56201_09011281823031_0221017801_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (407kB) | Request a copy
	Text RAMA_56201_09011281823031_0221017801_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (680kB) | Request a copy

Abstract

Cardiac arrest refers to a sudden interruption of cardiac activity that commonly caused by certain anomalous events. The patients of cardiac arrest have at least one abnormal vital sign in the one to four hours prior to the onset of cardiac arrest. In the previous studies, the patient's vital sign data have used many missing values. Due to the large number of missing values in the data, to process the data is so challenging. It is necessary to perform data imputation, in order to fill in the missing values in the patient's vital sign data. Machine learning for data imputation has been often implemented, but the result tends to get the poor performance with the datasets that have high missing values. Thus, deep learning methods are used, because they are proven to have the ability to explore and capture information hidden in data which makes progress in data imputation. This research proposes the Convolutional Neural Network (CNN) using three convolution layers and four convolution layers. CNN with three layers produces 88 models with the smallest RMSE result of 0.06378 and CNN with four layers of convolution produces 95 models with the smallest RMSE result of 0.062431.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Cardiac arrest, data imputation, missing value, convolutional neural network
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Thesa Jiornmia
Date Deposited:	15 Aug 2022 07:51
Last Modified:	15 Aug 2022 07:51
URI:	http://repository.unsri.ac.id/id/eprint/77263

Actions (login required)

View Item