Bilateral Ground Reaction Force Prediction Using Deep Learning Models and Custom Force Plate

• DOI: 10.31436/iiumej.v26i1.3379
• Publication Date: 2025-01-10T05:09:00Z
• Authors: Ying Heng Yeo; Muhammad Fauziniz...; Zaidi Mohd Ripin; Nur-Akasyah J.; Mohamad Ikhwan Za...; Alexander Wai Ten...; Jia Yi Tay
• Keywords: Ground reaction force

Several low-cost force plates have been proposed as alternatives for laboratory-grade plates. Nevertheless, the inability to quantify bilateral ground reaction (GRF) prevents these inexpensive from being used biomechanical analysis and certain clinical metric acquisition. This study developed deep-learning models, such autoencoder U-net, predict GRF vertical measured using a custom plate during sit-to-stand, gait initialization, gait. Results indicated that U-net model, which utilized STFT input, performed best. In addition predicting mediolateral model accurately predicted anterior-posterior in test dataset, achieving high Pearson's correlation coefficient, coefficient of determination, intraclass values over 0.90, 0.79, 0.89, respectively. The demonstrated higher compared three related previous studies different methods six inferring GRF. results potential TFU measurement tool perform bio-mechanical analysis. ABSTRAK: Beberapa plat daya kos rendah telah dicadangkan sebagai alternatif kepada berkualiti makmal. Walau bagaimanapun, ketidakmampuan untuk mengukur reaksi tanah secara menghalang yang murah ini daripada digunakan analisis biomekanik dan pengambilan metrik klinikal tertentu. Kajian membangunkan pembelajaran mendalam, seperti meramalkan menegak diukur menggunakan khas semasa pergerakan duduk-ke-berdiri, permulaan berjalan, berjalan. Hasil menunjukkan bahawa memberikan prestasi terbaik. Selain juga dengan tepat berjalan dalam set data ujian, mencapai nilai koefisien korelasi Pearson, penentuan, intrakelas tinggi melebihi masing-masing. Model Pearson lebih berbanding tiga kajian terdahulu berkaitan kaedah berbeza enam menyimpulkan mediolateral. potensi alat pengukuran melakukan biomekanik.