Rancang Bangun Sinyal ECG dan EEG Sebagai Deteksi Stres Pada Manusia Menggunakan Artificial Neural Network

Gondowijoyo, Steven Matthew (2023) Rancang Bangun Sinyal ECG dan EEG Sebagai Deteksi Stres Pada Manusia Menggunakan Artificial Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07311940000016-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (5MB) | Request a copy

Abstract

Emosi adalah keadaan pikiran tanpa disertai perubahan fisiologis. Emosi ini dapat dibagi menjadi emosi positif dan emosi negatif. Telah terbukti bahwa stres akut dan kronis mengubah beberapa variabel hemodinamik dan hormon seperti Heart Rate Variability (HRV) dan Sympathetic Nervous System (SNS) yang dapat digunakan sebagai ambang stres. Meningkatnya level stres ini dapat mendorong seseorang ke gejala penyakit mental yang kompleks dan menjadi faktor risiko yang signifikan untuk berbagai penyakit. Oleh karena itu, hal ini pendeteksian stres dengan tanda-tanda fisiologis dari tubuh manusia juga dilakukan, yaitu Electrocardiograph (ECG) dan Electroencephalography (EEG) yang terhubung dengan Mikromedia 7 FPI Capacitive dan terdapat chipset STM32F746ZG. Pengambilan sinyal ECG menggunakan aturan segitiga Einthoven dan pengambilan sinyal EEG pada titik Fp1 dan F3. Parameter yang dilakukan peninjauan berupa sinyal ECG dalam time domain, frequency domain, dan non-linear analysis, serta sinyal EEG dalam frequency domain. Pengujian dilakukan pada beberapa subjek dengan kondisi sehat berusia 18-23 tahun yang melakukan beberapa tahapan untuk menginduksi stres, dengan validasi berupa kuisioner STAI-Y1. Fitur HRV dan EEG tersebut memasuki proses pearson’s correlation function (PCF) yang mana menjadi input di metode klasifikasi. Dengan demikian, metode klasifikasi yang diajukan yaitu dengan Artificial Neural Network (ANN) dengan Stratified K-Fold memberikan keluaran tingkat stres sebesarr 95% dan nilai akurasi saat testing sebesar 97%. Proses evaluasi terhadap hasil kuesioner DASS21 dengan peninjauan tingkat stres secara objektif memberikan keluaran tingkat yang serupa yaitu sebanyak 72,73% dari sample yang dimiliki sedangkan hasil kuisioner STAI-Y1 memberikan nilai kesamaan sebanyak 90,91%. Penelitian dapat diaplikasikan kepada manusia yang terindikasi stres.
=================================================================================================================================
Emotion is a mental state without accompanying physiological changes. These emotions can be divided into positive and negative emotions. The bodily reactions associated with negative emotions are exemplified by stress. It has been proven that acute and chronic stress alters several hemodynamic and hormonal variables, such as Heart Rate Variability (HRV) and the Sympathetic Nervous System (SNS), which can be used as stress indicators. The elevation of stress levels can push an individual towards complex mental illnesses and serve as a significant risk factor for various diseases. Therefore, the detection of stress through physiological signs from the human body is also conducted, namely Electrocardiography (ECG) and Electroencephalography (EEG) that are connected to the Mikromedia 7 FPI Capacitive with the STM32F746ZG chipset. ECG signals are obtained using Einthoven's triangle rule, and EEG signals are obtained at the Fp1 and F3 points. The parameters under investigation include ECG signals in the time domain, frequency domain, and non-linear analysis, as well as EEG signals in the frequency domain. Testing is performed on several healthy subjects aged 18-23 years who undergo several stages to induce stress, with validation through the STAI-Y1 questionnaire. The HRV and EEG features undergo pearson's correlation function (PCF) process, which serves as input to the classification method. Hence, the proposed classification method is the Artificial Neural Network (ANN) with Stratified K-Fold, providing a stress level output of 95% and a testing accuracy of 97%. The evaluation process of the DASS21 questionnaire results, when compared to the objective assessment of stress levels, yields a similarity output of 72.73% from the sample population. Meanwhile, the STAI-Y1 questionnaire results yield a similarity score of 90.91%. This research can be applied to individuals who are indicated to be experiencing stress.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Artificial Neural Network, Deteksi Stres, ECG, EEG, Stratified K-Fold, Stress Detection.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76 Computer software > QA76.8 Microprocessor Q Science > QP Physiology > Q376.5 Electroencephalography (EEG) R Medicine > R Medicine (General) > R858 Deep Learning R Medicine > RC Internal medicine > RC683.5.E5 Electrocardiography T Technology > T Technology (General) > T58.62 Decision support systems T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Steven Matthew Gondowijoyo
Date Deposited:	01 Sep 2023 06:00
Last Modified:	01 Sep 2023 06:00
URI:	http://repository.its.ac.id/id/eprint/101205

Actions (login required)

View Item