Forecasting-based Biomedical Time-series Data Synthesis for Open Data and Robust AI

2025/10/06 by Y. Lee, Lee, Youngjoon, S. Cho +7
Computer Science · Neuroscience · #EEG and Brain-Computer Interfaces #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Machine Learning in Healthcare #Signal Processing (eess.SP) #Time Series Analysis and Forecasting #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2510.04622

openalex publication_date 2025/10/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The limited data availability due to strict privacy regulations and significant resource demands severely constrains biomedical time-series AI development, which creates a critical gap between data requirements and accessibility. Synthetic data generation presents a promising solution by producing artificial datasets that maintain the statistical properties of real biomedical time-series data without compromising patient confidentiality. While GANs, VAEs, and diffusion models capture global data distributions, forecasting models offer inductive biases tailored for sequential dynamics. We propose a framework for synthetic biomedical time-series data generation based on recent forecasting models that accurately replicates complex electrophysiological signals such as EEG and EMG with high fidelity. These synthetic datasets can be freely shared for open AI development and consistently improve downstream model performance. Numerical results on sleep-stage classification show up to a 3.71% performance gain with augmentation and a 91.00% synthetic-only accuracy that surpasses the real-data-only baseline.

Citations

• KAN: Kolmogorov-Arnold Networks
• SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion
• Structured Evaluation of Synthetic Tabular Data
• Time Weaver: A Conditional Time Series Generation Model
• Towards Generalizable Tumor Synthesis
• Universal Time-Series Representation Learning: A Survey
• iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
• Long-term Forecasting with TiDE: Time-series Dense Encoder
• Vector Quantized Time Series Generation with a Bidirectional Prior Model
• A Time Series is Worth 64 Words: Long-term Forecasting with Transformers
• TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis
• Are Transformers Effective for Time Series Forecasting?
• N-HiTS: Neural Hierarchical Interpolation for Time Series Forecasting
• NHITS: Neural Hierarchical Interpolation for Time Series Forecasting
• Parameter Efficient Deep Probabilistic Forecasting
• TimeVAE: A Variational Auto-Encoder for Multivariate Time Series Generation
• CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation
• Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting
• Temporal Fusion Transformers for interpretable multi-horizon time series forecasting
• An Alternative Probabilistic Interpretation of the Huber Loss
• N-BEATS: Neural basis expansion analysis for interpretable time series forecasting
• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• Dilated Recurrent Neural Networks
• Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
• DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks
• DeepAR: Probabilistic forecasting with autoregressive recurrent networks
• Deep Residual Learning for Image Recognition

Related

• Synthetic Event Time Series Health Data Generation
• MIRA: Medical Time Series Foundation Model for Real-World Health Data
• Forecasting adverse surgical events using self-supervised transfer learning for physiological signals
• Leveraging Generic Time Series Foundation Models for EEG Classification
• Time-Series Analysis on Edge-AI Hardware for Healthcare Monitoring