Reddy Lauhitya, Anand Ketan, Kaushik Shoibolina, Rodrigo Corey, McKay J Lucas, Kesar Trisha M, Kwon Hyeokhyen
Department of Biomedical Engineering, Emory University and Georgia Institute of Technology, Atlanta, Georgia, United States of America.
Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States of America.
PLOS Digit Health. 2025 Sep 16;4(9):e0001004. doi: 10.1371/journal.pdig.0001004. eCollection 2025 Sep.
Accurate diagnosis of gait impairments is often hindered by subjective or costly assessment methods, with current solutions relying on either expensive multi-camera equipment or subjective clinical observation. There is a critical need for accessible, objective tools that can aid in gait assessment while preserving patient privacy. In this work, we present a mobile phone-based, privacy-preserving artificial intelligence (AI) system for classifying gait impairments that leverages a novel dataset of 743 videos capturing seven distinct gait types. The dataset consists of frontal and sagittal views of clinicians simulating normal gait and six types of pathological gait (circumduction, Trendelenburg, antalgic, crouch, Parkinsonian, and vaulting), recorded using standard mobile phone cameras. Our system achieved 86.5% accuracy using combined frontal and sagittal views, with sagittal views generally outperforming frontal views except for specific gait types like circumduction. Model feature importance analysis revealed that frequency-domain features and entropy measures were critical for classification performance. Specifically, lower limb keypoints proved most important for classification, aligning with clinical understanding of gait assessment. These findings demonstrate that mobile phone-based systems can effectively classify diverse gait types while preserving privacy through on-device processing. The high accuracy achieved using simulated gait data suggests their potential for rapid prototyping of gait analysis systems, though clinical validation with patient data remains necessary. This work represents a significant step toward accessible, objective gait assessment tools for clinical, community, and tele-rehabilitation settings.
步态障碍的准确诊断常常受到主观或昂贵评估方法的阻碍，目前的解决方案要么依赖昂贵的多摄像头设备，要么依赖主观的临床观察。迫切需要可访问的、客观的工具，既能辅助步态评估，又能保护患者隐私。在这项工作中，我们提出了一种基于手机的、保护隐私的人工智能（AI）系统，用于对步态障碍进行分类，该系统利用了一个包含743个视频的新数据集，这些视频捕捉了七种不同的步态类型。该数据集包括临床医生模拟正常步态以及六种病理性步态（环行步态、特伦德伦伯格步态、疼痛步态、蹲伏步态、帕金森步态和跳跃步态）的正面和矢状面视图，使用标准手机摄像头录制。我们的系统使用正面和矢状面视图组合的准确率达到了86.5%，除了环行步态等特定步态类型外，矢状面视图的表现通常优于正面视图。模型特征重要性分析表明，频域特征和熵测度对分类性能至关重要。具体而言，下肢关键点对分类最为重要，这与步态评估的临床理解一致。这些发现表明，基于手机的系统可以有效地对多种步态类型进行分类，同时通过设备上的处理保护隐私。使用模拟步态数据取得的高准确率表明了它们在步态分析系统快速原型制作方面具有潜力，不过仍需要使用患者数据进行临床验证。这项工作朝着为临床、社区和远程康复环境提供可访问的、客观的步态评估工具迈出了重要一步。
