Patient-Level Early Identification of Freezing of Gait Using a Streamlined Temporal Model Based on Inertial Measurement Unit Data

Corresponding author: Jianqiang Gao, jianqianggaohh@126.com; Li Li, liliiot@163.com

DOI: 10.12201/bmr.202607.00038

•  

  Abstract: Objective/Significance This study presents a wearable inertial measurement unit (IMU)-based framework for early freezing of gait (FOG) identification in previously unseen patients, specifically targeting cross-patient generalizability and dissecting the relative contributions of key spectral features and network modules. Methods/Process We interrogated a publicly available dataset comprising 56 recordings from 12 patients. After verifying sensor availability, we performed direction-invariant feature aggregation from shank-mounted IMU signals. A compact temporal architecture was subsequently designed, integrating residual one-dimensional convolutions, bidirectional long short-term memory (LSTM) networks, and an attention-based pooling mechanism. Internal validation employed a patient-level leave-one-subject-out (LOSO) scheme, alongside systematic ablation studies, all conducted across three fixed random seeds. Results/Conclusion The proposed pipeline yielded an accuracy of 0.7910, a balanced accuracy of 0.7378, an F1-score of 0.6325, an area under the receiver operating characteristic curve (AUROC) of 0.8791, and an area under the precision–recall curve (AUPRC) of 0.7595. The freeze- and gait-band spectral features, together with bidirectional temporal modeling, emerged as primary drivers of performance. Notably, across the three fixed seeds during LOSO internal validation on the current dataset, the AUROC and AUPRC exhibited only limited variability, confirming reliable cross-patient stability.

  Key words: Parkinson’s disease; freezing of gait; inertial measurement unit; early identification; patient-level cross-validation

  Submit time: 26 August 2026

  Copyright: The copyright holder for this preprint is the author/funder, who has granted biomedRxiv a license to display the preprint in perpetuity.
• html

• 图表

• li da. The diagnostic value of olfactory testing combined with multiple indicators for early Parkinsons disease. 2025. doi: 10.12201/bmr.202512.00075

  HUO Jie, HANG Feifei, XU Jing, GUAN Zheng. Value of serum IgE, LDH, and IL-18 in the early identification of refractory mycoplasma pneumoniae pneumonia in children. 2026. doi: 10.12201/bmr.202606.00009

  zhoudong, zhusiying, gaoxinyi, sunxibin, yuanbeibei. The relationship between job satisfaction of primary healthcare workers and patient satisfaction: the cross-level mediating effect of perceived quality of health care. 2026. doi: 10.12201/bmr.202606.00082

  ZHANG MIN, WU YINXIN. Research on the Measurement and Optimization Path of Coordination Degree in Collaborative Supervision of China’s Medical Insurance Fund—Based on the SFIC Framework. 2023. doi: 10.12201/bmr.202311.00006

  Li Huifang, Li Jianchun, Zhai Yaxian, Yang Jinmei, Shen Yunxia. Systematic Review of Factors Influencing Olfactory Dysfunction in Patients with Parkinsons Disease. 2025. doi: 10.12201/bmr.202503.00057

  liangleran, Xu Qian, Yang Meng, WuJiaheng, 陈振虎, Liuxiufeng. Intelligent Q&A Study of Traditional Chinese Medicine for Parkinsons Disease Based on Knowledge Graph and Large Language Modeling. 2026. doi: 10.12201/bmr.202604.00033

  HAN LEYANG, LIU Xiaoyun. Association between integrated service level and outpatient care-seeking patterns among rural chronic disease patients: a cross-sectional study in three Chinese counties. 2026. doi: 10.12201/bmr.202608.00036

  HE Lanlan, LI Danyang, SHEN Li, WU Zhonghua, ZHANG Jun, YE Yongqiang. Construction and validation of nomogram model for prolonged length of stay in patients with acute cerebral infarction based on total cerebral small vessel disease burden scores. 2025. doi: 10.12201/bmr.202501.00008

  mayila·abudukelimu. Application of AI tool in early prediction of secondary pulmonary hypertension in patients with COPD. 2026. doi: 10.12201/bmr.202607.00013

  SHU Wenxiu, LUO Liufei, Tong Jiaqi, Le Jing. Construction and effect evaluation of risk prediction model for grades 3-4 myeloma bone disease in newly diagnosed multiple myeloma patients. 2024. doi: 10.12201/bmr.202410.00057

• ID	Submit time	Number	Download
  1	2026-07-24	10.12201/bmr.202607.00038V1	Download

• Public  Anonymous  To author only Submit Comments