Open access Jurnal SINTA 2
Sistem Deteksi Stres Menggunakan Metode Support Vector Machine dengan Fitur RMSSD dan PNN50
Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK) · Vol. 13 No. 3 (2026) · pp. 703–712
- ISSN
- 2528-6579, 2355-7699
- Published
- 2026-06-30
- Publisher
- Fakultas Ilmu Komputer, Universitas Brawijaya
- Language
- id
Abstract
Bahasa Indonesia
Stres merupakan respons fisiologis dan psikologis yang muncul akibat tekanan fisik, lingkungan, atau sosial. Keterbatasan jumlah psikolog klinis berpotensi menyebabkan keterlambatan skrining dan penanganan dini, sehingga diperlukan alat bantu skrining yang objektif dan terukur untuk memperkuat pemeriksaan awal. Penelitian ini mengusulkan prototipe deteksi stres non-invasif berbasis embedded system secara real-time dengan memanfaatkan sensor PPG MAX30102, ekstraksi fitur HRV domain waktu RMSSD dan pNN50, serta klasifikasi menggunakan Support Vector Machine (SVM) linear pada dua kelas (stres/tidak stres). Data primer direkam dari subjek pada kondisi stres/tidak stres yang ditetapkan melalui validasi psikiater, kemudian diproses untuk memperoleh nilai RMSSD dan pNN50 sebelum dilakukan klasifikasi. Hasil pengujian menunjukkan sistem mampu melakukan estimasi detak jantung dengan MAPE 2,25%, sedangkan perhitungan NN interval sebagai dasar ekstraksi HRV menunjukkan MAPE 0,15%. Pada tahap deteksi stres, sistem mencapai akurasi 90% dengan waktu komputasi rata-rata 86,5 ms, sehingga layak digunakan sebagai alat bantu skrining stres awal yang praktis, portabel, dan responsif pada perangkat embedded. Abstract Stress is a physiological and psychological response that arises from physical, environmental, or social pressures. The limited number of clinical psychologists may lead to delays in early screening and intervention, highlighting the need for an objective and measurable screening aid to strengthen preliminary assessments. This study proposes a non-invasive, real-time embedded system prototype for stress detection using a MAX30102 PPG sensor, time-domain HRV feature extraction (RMSSD and pNN50), and a linear Support Vector Machine (SVM) classifier for two classes (stress/non-stress). Primary data were recorded from participants under stress and non-stress conditions validated by a psychiatrist, then processed to obtain RMSSD and pNN50 values prior to classification. The experimental results show that the system achieved heart rate estimation with a MAPE of 2.25%, while NN interval computation as the basis for HRV extraction achieved a MAPE of 0.15%. For stress detection, the system reached an accuracy of 90% with an average computation time of 86.5 ms, indicating that it is suitable as a practical, portable, and responsive early stress screening tool on an embedded platform.
English
Stres merupakan respons fisiologis dan psikologis yang muncul akibat tekanan fisik, lingkungan, atau sosial. Keterbatasan jumlah psikolog klinis berpotensi menyebabkan keterlambatan skrining dan penanganan dini, sehingga diperlukan alat bantu skrining yang objektif dan terukur untuk memperkuat pemeriksaan awal. Penelitian ini mengusulkan prototipe deteksi stres non-invasif berbasis embedded system secara real-time dengan memanfaatkan sensor PPG MAX30102, ekstraksi fitur HRV domain waktu RMSSD dan pNN50, serta klasifikasi menggunakan Support Vector Machine (SVM) linear pada dua kelas (stres/tidak stres). Data primer direkam dari subjek pada kondisi stres/tidak stres yang ditetapkan melalui validasi psikiater, kemudian diproses untuk memperoleh nilai RMSSD dan pNN50 sebelum dilakukan klasifikasi. Hasil pengujian menunjukkan sistem mampu melakukan estimasi detak jantung dengan MAPE 2,25%, sedangkan perhitungan NN interval sebagai dasar ekstraksi HRV menunjukkan MAPE 0,15%. Pada tahap deteksi stres, sistem mencapai akurasi 90% dengan waktu komputasi rata-rata 86,5 ms, sehingga layak digunakan sebagai alat bantu skrining stres awal yang praktis, portabel, dan responsif pada perangkat embedded. Abstract Stress is a physiological and psychological response that arises from physical, environmental, or social pressures. The limited number of clinical psychologists may lead to delays in early screening and intervention, highlighting the need for an objective and measurable screening aid to strengthen preliminary assessments. This study proposes a non-invasive, real-time embedded system prototype for stress detection using a MAX30102 PPG sensor, time-domain HRV feature extraction (RMSSD and pNN50), and a linear Support Vector Machine (SVM) classifier for two classes (stress/non-stress). Primary data were recorded from participants under stress and non-stress conditions validated by a psychiatrist, then processed to obtain RMSSD and pNN50 values prior to classification. The experimental results show that the system achieved heart rate estimation with a MAPE of 2.25%, while NN interval computation as the basis for HRV extraction achieved a MAPE of 0.15%. For stress detection, the system reached an accuracy of 90% with an average computation time of 86.5 ms, indicating that it is suitable as a practical, portable, and responsive early stress screening tool on an embedded platform.
Keywords
Collaboration network of the authors
21 authors · 45 co-authorship links
Hover or tap a name to highlight its co-authors. Double-click to open the author's page.
- University of Brawijaya
- Affiliation unknown
- Different organisations
- Same organisation
- Affiliation unknown
- Thicker line = more joint articles · bigger circle = more articles
Show as table
| Co-author | Organisation | Joint articles |
|---|---|---|
| Barlian Henryranu Prasetio | University of Brawijaya | 3 |
| Eko Setiawan | University of Brawijaya | 2 |
| Hurriyatul Fitriyah | University of Brawijaya | 1 |
| Ryzaldi Ananda Fabiana | University of Brawijaya | 1 |
| Syahrul Chilmi | University of Brawijaya | 1 |
| Dahnial Syauqy | University of Brawijaya | 1 |
| Septiyo Budi Perkasa | University of Brawijaya | 1 |
| Rizal Maulana | University of Brawijaya | 1 |
| Dittha Ratanasari | University of Brawijaya | 1 |
| Agung Setia Budi | University of Brawijaya | 1 |
| Ragil Hadi Prasetyo | University of Brawijaya | 1 |
| Achmad Ghifari | University of Brawijaya | 1 |
| Gabriel Hizkia Marhata Simbolon | Affiliation unknown | 1 |
| Gabriel Hizkia Marhata Simbolon | Affiliation unknown | 1 |
| Teguh Surya Darma Santoso | Affiliation unknown | 1 |
| Syifa’ Hukma Shabiyya | Affiliation unknown | 1 |
| Muhammad Rava Athalla | Affiliation unknown | 1 |
| Alifa Azwadina Fuad | Affiliation unknown | 1 |
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.