Department of Systems Design Engineering, University of Waterloo, Waterloo, ON, Canada.
Faculty of Medicine, University of Ottawa and Ottawa Hospital Research Institute, Ottawa, ON, Canada.
J Neuroeng Rehabil. 2021 Nov 27;18(1):167. doi: 10.1186/s12984-021-00958-5.
Freezing of gait (FOG) is a walking disturbance in advanced stage Parkinson's disease (PD) that has been associated with increased fall risk and decreased quality of life. Freezing episodes can be mitigated or prevented with external intervention such as visual or auditory cues, activated by FOG prediction and detection systems. While most research on FOG detection and prediction has been based on inertial measurement unit (IMU) and accelerometer data, plantar-pressure data may capture subtle weight shifts unique to FOG episodes. Different machine learning algorithms have been used for FOG detection and prediction; however, long short-term memory (LSTM) deep learning methods hold an advantage when dealing with time-series data, such as sensor data. This research aimed to determine if LSTM can be used to detect and predict FOG from plantar pressure data alone, specifically for use in a real-time wearable system.
Plantar pressure data were collected from pressure-sensing insole sensors worn by 11 participants with PD as they walked a predefined freeze-provoking path. FOG instances were labelled, 16 features were extracted, and the dataset was balanced and normalized (z-score). The resulting datasets were classified using long short-term memory neural-network models. Separate models were trained for detection and prediction. For prediction models, data before FOG were included in the target class. Leave-one-freezer-out cross validation was used for model evaluation. In addition, the models were tested on all non-freezer data to determine model specificity.
The best FOG detection model had 82.1% (SD 6.2%) mean sensitivity and 89.5% (SD 3.6%) mean specificity for one-freezer-held-out cross validation. Specificity improved to 93.3% (SD 4.0%) when ignoring inactive state data (standing) and analyzing the model only on active states (turning and walking). The model correctly detected 95% of freeze episodes. The best FOG prediction method achieved 72.5% (SD 13.6%) mean sensitivity and 81.2% (SD 6.8%) mean specificity for one-freezer-held-out cross validation.
Based on FOG data collected in a laboratory, the results suggest that plantar pressure data can be used for FOG detection and prediction. However, further research is required to improve FOG prediction performance, including training with a larger sample of people who experience FOG.
冻结步态（FOG）是一种在帕金森病（PD）晚期出现的行走障碍，与增加跌倒风险和降低生活质量有关。通过冻结预测和检测系统，外部干预（如视觉或听觉提示）可以减轻或预防冻结发作。虽然大多数关于 FOG 检测和预测的研究都是基于惯性测量单元（IMU）和加速度计数据，但足底压力数据可能会捕捉到与 FOG 发作独特的微妙体重转移。不同的机器学习算法已被用于 FOG 检测和预测；然而，长短期记忆（LSTM）深度学习方法在处理时间序列数据（如传感器数据）方面具有优势。本研究旨在确定 LSTM 是否可单独用于从足底压力数据中检测和预测 FOG，特别是用于实时可穿戴系统。
11 名 PD 患者穿着压力感应鞋垫传感器行走预先设定的诱发冻结路径时，收集足底压力数据。FOG 实例被标记，提取了 16 个特征，并对数据集进行了平衡和标准化（z 分数）。使用长短期记忆神经网络模型对分类后的数据集进行分类。分别为检测和预测训练模型。对于预测模型，将 FOG 之前的数据包含在目标类中。使用留一冻结者交叉验证进行模型评估。此外，还在所有非冻结者数据上测试模型，以确定模型特异性。
对于一冻结者留一交叉验证，最佳 FOG 检测模型的平均灵敏度为 82.1%（SD 6.2%），平均特异性为 89.5%（SD 3.6%）。当忽略非活动状态（站立）数据并仅在活动状态（转弯和行走）上分析模型时，特异性提高到 93.3%（SD 4.0%）。模型正确检测到 95%的冻结发作。最佳 FOG 预测方法的平均灵敏度为 72.5%（SD 13.6%），平均特异性为 81.2%（SD 6.8%），对于一冻结者留一交叉验证。
基于在实验室中收集的 FOG 数据，结果表明足底压力数据可用于 FOG 检测和预测。然而，需要进一步研究以提高 FOG 预测性能，包括使用经历 FOG 的更大人群样本进行训练。
