Items where Author is "Aripin, Moch."

Aripin, Moch. and Hardhienata, Soewarto and Puja Negara, Teguh (2023) J-INTECH (Journal of Information and Technology) Terakreditasi Kemendikbud SK No. 204/E/KPT/2022 E-ISSN: 2580-720X || P-ISSN: 2303-1425 Deteksi Gangguan Tidur Menggunakan Metode Jaringan Saraf Tiruan Berbasis Internet Of Things (IOT) Moch Aripin1* , Soewarto Hardhienata2, Teguh Puja Negara3 1,2,3Universitas Pakuan, Fakultas Matematika dan Ilmu Pengetahuan Alam, Program Studi Ilmu Komputer, Bogor, Indonesia Informasi Artikel Abstrak Diterima: 26-05-2024 Direvisi: 27-05-2024 Diterbitkan: 28-06-2024 Gangguan tidur seperti Central Sleep Apnea (CSA) dan Obstructive Sleep Apnea (OSA) dapat berdampak buruk bagi kesehatan jika tidak ditangani dengan baik. Penelitian ini bertujuan merancang alat pendeteksi gangguan tidur menggunakan metode jaringan saraf tiruan berbasis Internet of Things (IoT). Sistem ini menggunakan sensor AD8232 untuk mengakuisisi sinyal elektrokardiogram (EKG) yang kemudian diekstraksi fitur High Frequency dan Low Frequency. Ekstraksi fitur dilakukan dengan metode Fast Fourier Transform. Klasifikasi kondisi normal, CSA, atau OSA dilakukan dengan metode jaringan saraf tiruan Multilayer Perceptron yang ditraining menggunakan data dari Physionet. Mikrokontroler ESP32 digunakan untuk memproses ekstraksi fitur dan klasifikasi. Hasil klasifikasi kemudian dikirimkan ke database melalui modul WiFi ESP32 dan ditampilkan pada antarmuka website. Dari pengujian kinerja sensor AD8232 diperoleh akurasi 96,85%, akurasi klasifikasi menggunakan Jaringan Syaraf Tiruan sebesar 80%, dan waktu komputasi rata-rata 7,6 ms. Sistem ini berpotensi membantu deteksi dini gangguan tidur sehingga dapat ditangani lebih awal oleh tenaga medis. Kata Kunci Gangguan tidur; Central Sleep Apnea; obstructive Sleep Apnea; Jaringan Saraf Tiruan; Internet of Things *Email Korespondensi: mocharipin214@gmail.com Abstract Sleep disorders such as Central Sleep Apnea (CSA) and Obstructive Sleep Apnea (OSA) can have adverse health effects if not treated properly. This research aims to design a sleep disorder detection device using the Internet of Things (IoT)- based artificial neural network method. This system uses AD8232 sensor to acquire electrocardiogram (ECG) signal which is then extracted High Frequency and Low Frequency features. Feature extraction is performed using the Fast Fourier Transform method. Classification of normal, CSA, or OSA conditions is performed using the Multilayer Perceptron artificial neural network method which is trained using data from Physionet. The ESP32 microcontroller is used to process the feature extraction and classification. The classification results are then sent to the database via the ESP32 WiFi module and displayed on the website interface. From testing the performance of the AD8232 sensor, an accuracy of 96.85% was obtained, the classification accuracy using the Artificial Neural Network was 80%, and the average computation time was 7.6 ms. This system has the potential to help early detection of sleep disorders so that they can be treated early by medical personnel. Skripsi thesis, Universitas Pakuan.