SISTEM PENDETEKSI TINGKAT KECEMASAN PADA PENDERITA ASMA MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK (ANN)

Imam, Muhammad Allif (2024) SISTEM PENDETEKSI TINGKAT KECEMASAN PADA PENDERITA ASMA MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK (ANN). Diploma thesis, Universitas Andalas.

	Text (Cover dan Abstrak) cover-2011512001.pdf - Published Version Download (475kB)
	Text (Bab 1 Pendahuluan) BAB 1-2011512001.pdf - Published Version Download (751kB)
	Text (Bab 5 Penutup) BAB V-2011512001.pdf - Published Version Download (438kB)
	Text (Daftar Pustaka) DAFTAR PUSTAKA-2011512001.pdf - Published Version Download (695kB)
	Text (skripsi full text) Laporan TA IMAM.pdf - Published Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Anxiety detection in asthma patients is a critical aspect of managing their condition. This research aims to design a system capable of detecting emotional states, particularly anxiety, in asthma patients using the Artificial Neural Network (ANN) method. The system is designed to raise awareness among those around the patient and provide accurate anxiety detection. The test results show that the system can measure heart rate with an average deviation of 1.4 and an average error of 1.7%, achieving an accuracy of 98.3% compared to a pulse oximeter. Additionally, the system can measure skin conductivity with an average deviation of 0.496 and an average error of 8.8%, achieving an accuracy of 91.2% compared to a digital multimeter. The system successfully classified anxiety levels according to the parameters, with the ANN architecture using 2 hidden layers achieving a test accuracy of 0.78%, outperforming the architecture with 1 hidden layer, which had an accuracy of 0.76%. However, the implementation on Arduino Uno revealed limitations in supporting the ANN architecture with 2 hidden layers.

Actions (login required)

View Item