Debnath Minakshi, Alamgeer Sana, Kabir Md Shahriar, Ngu Anne H
Department of Computer Science, Texas State University, San Marcos, TX 78666-4684, USA.
Sensors (Basel). 2025 Jul 26;25(15):4639. doi: 10.3390/s25154639.
Deep learning models rely heavily on extensive training data, but obtaining sufficient real-world data remains a major challenge in clinical fields. To address this, we explore methods for generating realistic synthetic multivariate fall data to supplement limited real-world samples collected from three fall-related datasets: SmartFallMM, UniMib, and K-Fall. We apply three conventional time-series augmentation techniques, a Diffusion-based generative AI method, and a novel approach that extracts fall segments from public video footage of older adults. A key innovation of our work is the exploration of two distinct approaches: video-based pose estimation to extract fall segments from public footage, and Diffusion models to generate synthetic fall signals. Both methods independently enable the creation of highly realistic and diverse synthetic data tailored to specific sensor placements. To our knowledge, these approaches and especially their application in fall detection represent rarely explored directions in this research area. To assess the quality of the synthetic data, we use quantitative metrics, including the Fréchet Inception Distance (FID), Discriminative Score, Predictive Score, Jensen-Shannon Divergence (JSD), and Kolmogorov-Smirnov (KS) test, and visually inspect temporal patterns for structural realism. We observe that Diffusion-based synthesis produces the most realistic and distributionally aligned fall data. To further evaluate the impact of synthetic data, we train a long short-term memory (LSTM) model offline and test it in real time using the SmartFall App. Incorporating Diffusion-based synthetic data improves the offline F1-score by 7-10% and boosts real-time fall detection performance by 24%, confirming its value in enhancing model robustness and applicability in real-world settings.
深度学习模型严重依赖大量的训练数据，但在临床领域获取足够的真实世界数据仍然是一项重大挑战。为了解决这一问题，我们探索了生成逼真的多变量合成跌倒数据的方法，以补充从三个与跌倒相关的数据集（SmartFallMM、UniMib和K-Fall）收集的有限真实世界样本。我们应用了三种传统的时间序列增强技术、一种基于扩散的生成式人工智能方法，以及一种从老年人公共视频片段中提取跌倒片段的新颖方法。我们工作的一个关键创新点是探索了两种不同的方法：基于视频的姿态估计从公共片段中提取跌倒片段，以及扩散模型生成合成跌倒信号。这两种方法都能够独立创建针对特定传感器放置的高度逼真且多样的合成数据。据我们所知，这些方法，尤其是它们在跌倒检测中的应用，在该研究领域代表了很少被探索的方向。为了评估合成数据的质量，我们使用了定量指标，包括弗雷歇因距离（FID）、判别分数、预测分数、詹森 - 香农散度（JSD）和柯尔莫哥洛夫 - 斯米尔诺夫（KS）检验，并直观地检查时间模式以确保结构逼真性。我们观察到基于扩散的合成产生了最逼真且分布最匹配的跌倒数据。为了进一步评估合成数据的影响，我们离线训练一个长短期记忆（LSTM）模型，并使用SmartFall应用程序进行实时测试。纳入基于扩散的合成数据将离线F1分数提高了7 - 10%，并将实时跌倒检测性能提高了24%，证实了其在增强模型鲁棒性和在现实世界环境中的适用性方面的价值。
