Klasifikasi Skor Action Research Arm Test Menggunakan Wearable sensor Berbasis IoT dan Random Forest untuk Rehabilitasi Pasca Stroke
DOI:
https://doi.org/10.52436/1.jpti.2199Kata Kunci:
Action Research Arm Test, Internet of Things, Random Forest, Rehabilitasi Pascastroke, Wearable sensorAbstrak
Pemantauan fungsi ekstremitas atas merupakan bagian penting dalam rehabilitasi pascastroke, tetapi penilaian secara konvensional masih bergantung pada pengamatan tenaga kesehatan dan sulit dilakukan secara berkelanjutan. Penelitian ini bertujuan mengembangkan prototipe hand wearable sensor berbasis Internet of Things untuk mengklasifikasikan kualitas gerakan tangan berdasarkan skor Action Research Arm Test (ARAT). Perangkat dibangun menggunakan mikrokontroler ESP32 dan dua sensor MPU6050 yang ditempatkan pada punggung tangan dan lengan bawah untuk merekam data akselerometer dan giroskop dari 10 subjek sehat pada tujuh gerakan ARAT yang mencakup kategori gross movement dan grip. Fitur statistik diekstraksi dari sinyal sensor dan ketidakseimbangan data pada setiap kelas ditangani sebelum tujuh model Random Forest dikembangkan secara terpisah untuk memprediksi skor ARAT 0–3 pada setiap gerakan. Hasil pengujian menunjukkan rata-rata akurasi sebesar 89%, presisi makro 0,83, recall makro 0,84, dan F1-score makro 0,84, dengan model gerakan tangan ke belakang kepala mencapai performa tertinggi (akurasi 94%, F1-score 0,90). Hasil tersebut menunjukkan bahwa sistem mampu mengenali pola kualitas gerakan secara menjanjikan sebagai prototipe pendukung pemantauan rehabilitasi, meskipun validasi klinis pada pasien pascastroke masih diperlukan sebelum sistem digunakan dalam pelayanan kesehatan.
Unduhan
Referensi
A. A. Zainuddin et al., “The major risk factor of stroke across Indonesia; a nationwide geospatial analysis of universal health coverage program,” Arch. Public Heal., vol. 83, no. 1, 2025, doi: 10.1186/s13690-025-01613-4.
A. Bhalla, L. Clark, R. Fisher, and M. James, “The new national clinical guideline for stroke: an opportunity to transform stroke care,” Clin. Med. (Northfield. Il)., vol. 24, no. 2, p. 100025, 2024, doi: 10.1016/j.clinme.2024.100025.
L. Daghsen et al., “Evaluation of a shortened version of the Action Research Arm Test (ARAT) for upper extremity function after stroke: The Mini-ARAT,” Clin. Rehabil., vol. 36, no. 9, pp. 1257–1266, 2022, doi: 10.1177/02692155221097322.
J. Fernández-Solana et al., “Psychometric Properties of the Action Research Arm Test (ARAT) Scale in Post-Stroke Patients—Spanish Population,” Int. J. Environ. Res. Public Health, vol. 19, no. 22, p. 14918, 2022, doi: 10.3390/ijerph192214918.
L. Zhou et al., “Monitoring and Visualizing Stroke Rehabilitation Progress using Wearable Sensors,” 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, pp. 1–6, 2024. doi: 10.1109/embc53108.2024.10782489.
P. Maceira-Elvira, T. Popa, A.-C. Schmid, and F. C. Hummel, “Wearable technology in stroke rehabilitation: towards improved diagnosis and treatment of upper-limb motor impairment,” J. Neuroeng. Rehabil., vol. 16, no. 1, 2019, doi: 10.1186/s12984-019-0612-y.
C. Werner et al., “Using Wearable Inertial Sensors to Estimate Clinical Scores of Upper Limb Movement Quality in Stroke,” Front. Physiol., vol. 13, 2022, doi: 10.3389/fphys.2022.877563.
I. Boukhennoufa, X. Zhai, V. Utti, J. Jackson, and K. D. McDonald-Maier, “Wearable sensors and machine learning in post-stroke rehabilitation assessment: A systematic review,” Biomed. Signal Process. Control, vol. 71, p. 103197, 2022, doi: 10.1016/j.bspc.2021.103197.
J. Ukey, C. Rogers, S. Uhlrich, M. Akcakaya, and A. Sethi, “International Journal of Medical Informatics Enhancing stroke recovery assessment?: A machine learning approach to real-world hand function analysis,” Int. J. Med. Inform., vol. 204, no. June, p. 106077, 2025, doi: 10.1016/j.ijmedinf.2025.106077.
C. Yang et al., “Perspectives of users for a future interactive wearable system for upper extremity rehabilitation following stroke: a qualitative study,” J. Neuroeng. Rehabil., vol. 20, no. 1, 2023, doi: 10.1186/s12984-023-01197-6.
S. Selviana, M. Subito, R. Fauzi, and A. Alamsyah, “RANCANG BANGUN ALAT MONITORING PERKEMBANGAN PASIEN PASCA STROKE BERBASIS IoT (INTERNET of THINGS),” Foristek, vol. 11, no. 2, 2021, doi: 10.54757/fs.v11i2.107.
K. Boardsworth et al., “Upper limb robotic rehabilitation following stroke: a systematic review and meta-analysis investigating efficacy and the influence of device features and program parameters,” J. Neuroeng. Rehabil., vol. 22, no. 1, 2025, doi: 10.1186/s12984-025-01662-4.
M. Fikry, Bustami, and E. Suzanna, “IOT-Based Environmental Monitoring and Its Implications For Student Mental Health,” Rev. Gestão Soc. e Ambient., vol. 19, no. 2, p. e011183, 2025, doi: 10.24857/rgsa.v19n2-022.
Y. Susanthi, A. Darmawan, and S. B. Suryadi, “Alat Pengenal Gerakan Tangan Menggunakan Algoritma Artificial Neural Network Berbasis ESP32 dan Platform Edge Impulse,” Techné J. Ilm. Elektrotek., vol. 23, no. 1, pp. 77–94, 2024, doi: 10.31358/techne.v23i1.434.
N. Alhammad, “applied sciences Dynamic Segmentation for Physical Activity Recognition Using a Single Wearable Sensor,” 2021.
T. Accelerometers, M. Bennasar, B. A. Price, D. Gooch, and A. K. Bandara, “Significant Features for Human Activity Recognition Using,” 2022.
M. Li et al., “Application of Deep Learning Algorithm to Monitor Upper Extremity Task Practice,” pp. 1–16, 2023.
