arXiv subjects

Warren Woodrich Pettine

Publications and source records attributed to Warren Woodrich Pettine.

4 recordsLinked to original sources

arXiv2026

Timesynth: A Temporal Fidelity Framework for Health Signal Digital Twins

Forecasting models for health-signal digital twins must preserve the oscillatory, frequency, phase, and state-transition dynamics of physiological signals, yet the pointwise metrics used to benchmark them cannot detect when these fundamental properties are lost. We show that this blind spot misranks models: across 11 architectures, models with comparable pointwise error diverge by up to 53° in phase accuracy, equivalent to roughly 123 ms for a 1.2 Hz cardiac rhythm and invisible to standard metrics. To enable development of models that escape such failures, we introduce TimeSynth, a controlled benchmarking framework with two reusable components: a physiologically grounded generator producing signals with analytically known ground-truth dynamics from parametric models fitted to real electroencephalography, electrocardiography and photoplethysmogram signals, along with diagnostics quantifying amplitude, frequency, phase, and state-transition fidelity. Linear and full-sequence attention models systematically lose frequency and phase information despite acceptable amplitude error, whereas architectures with localized temporal structure better preserve dynamical fidelity and adapt to observable state transitions; none, however, reliably preserves stochastic switching. Because the dominant determinant of fidelity is architectural, model choice becomes a principled, use-case-driven decision rather than a search for a single winner. TimeSynth thus supplies the controlled preclinical stress test missing before models are coupled to patient data, with a reusable generator and diagnostics for fidelity-aware development.

cs.LG↗

arXiv2026

TimeSynth: A Framework for Uncovering Systematic Biases in Time Series Forecasting

Time series forecasting is a fundamental tool with wide ranging applications, yet recent debates question whether complex nonlinear architectures truly outperform simple linear models. Prior claims of dominance of the linear model often stem from benchmarks that lack diverse temporal dynamics and employ biased evaluation protocols. We revisit this debate through TimeSynth, a structured framework that emulates key properties of real world time series,including non-stationarity, periodicity, trends, and phase modulation by creating synthesized signals whose parameters are derived from real-world time series. Evaluating four model families Linear, Multi Layer Perceptrons (MLP), Convolutional Neural Networks (CNNs), and Transformers, we find a systematic bias in linear models: they collapse to simple oscillation regardless of signal complexity. Nonlinear models avoid this collapse and gain clear advantages as signal complexity increases. Notably, Transformers and CNN based models exhibit slightly greater adaptability to complex modulated signals compared to MLPs. Beyond clean forecasting, the framework highlights robustness differences under distribution and noise shifts and removes biases of prior benchmarks by using independent instances for train, test, and validation for each signal family. Collectively, TimeSynth provides a principled foundation for understanding when different forecasting approaches succeed or fail, moving beyond oversimplified claims of model equivalence.

cs.LG↗

arXiv2025

A Practical SAFE-AI Framework for Small and Medium-Sized Enterprises Developing Medical Artificial Intelligence Ethics Policies

Artificial intelligence (AI) offers incredible possibilities for patient care, but raises significant ethical issues, such as the potential for bias. Powerful ethical frameworks exist to minimize these issues, but are often developed for academic or regulatory environments and tend to be comprehensive but overly prescriptive, making them difficult to operationalize within fast-paced, resource-constrained environments. We introduce the Scalable Agile Framework for Execution in AI (SAFE-AI) designed to balance ethical rigor with business priorities by embedding ethical oversight into standard Agile-based product development workflows. The framework emphasizes the early establishment of testable acceptance criteria, fairness metrics, and transparency metrics to manage model uncertainty, while also promoting continuous monitoring and re-evaluation of these metrics across the AI lifecycle. A core component of this framework are responsibility metrics using scenario-based probability analogy mapping designed to enhance transparency and stakeholder trust. This ensures that retraining or tuning activities are subject to lightweight but meaningful ethical review. By focusing on the minimum necessary requirements for responsible development, our framework offers a scalable, business-aligned approach to ethical AI suitable for organizations without dedicated ethics teams.

cs.CY↗

arXiv2024

Assessing Foundation Models' Transferability to Physiological Signals in Precision Medicine

The success of precision medicine requires computational models that can effectively process and interpret diverse physiological signals across heterogeneous patient populations. While foundation models have demonstrated remarkable transfer capabilities across various domains, their effectiveness in handling individual-specific physiological signals - crucial for precision medicine - remains largely unexplored. This work introduces a systematic pipeline for rapidly and efficiently evaluating foundation models' transfer capabilities in medical contexts. Our pipeline employs a three-stage approach. First, it leverages physiological simulation software to generate diverse, clinically relevant scenarios, particularly focusing on data-scarce medical conditions. This simulation-based approach enables both targeted capability assessment and subsequent model fine-tuning. Second, the pipeline projects these simulated signals through the foundation model to obtain embeddings, which are then evaluated using linear methods. This evaluation quantifies the model's ability to capture three critical aspects: physiological feature independence, temporal dynamics preservation, and medical scenario differentiation. Finally, the pipeline validates these representations through specific downstream medical tasks. Initial testing of our pipeline on the Moirai time series foundation model revealed significant limitations in physiological signal processing, including feature entanglement, temporal dynamics distortion, and reduced scenario discrimination. These findings suggest that current foundation models may require substantial architectural modifications or targeted fine-tuning before deployment in clinical settings.

cs.LG↗