Back to Evidence Hub
Case Study seismocardiography-pig-hypovolemia-dataset-for-signal-quality-indexing-and-validated-cardiac-timings
2026 Release

Seismocardiography Pig Hypovolemia Dataset for Signal Quality Indexing and Validated Cardiac Timings

Executive Summary

This study introduces a rigorously annotated dataset of 17,059 seismocardiography (SCG) heartbeats collected from six porcine subjects undergoing hypovolemia protocols. Using a custom graphical user interface (GUI), expert annotators identified fiducial points (e.g., aortic opening and closing) and signal quality indices (SQI), achieving high inter-annotator agreement and strong correlations with gold-standard catheter-based measurements (r=0.926 for AO and r=0.911 for AC). The dataset addresses the scarcity of validated SCG data, enabling advancements in real-time cardiac monitoring, algorithm development, and diagnostic applications.

This study provides a high-quality dataset of heart vibrations from pigs, helping researchers develop better tools for tracking heart health using wearable sensors.

Answer Machine Insights

Q: What is the primary contribution of this dataset?

The dataset provides rigorously annotated SCG signals with validated fiducial points and signal quality indices, addressing the lack of standardized SCG datasets.

Annotated SCG dataset supports advancements in real-time cardiac monitoring, denoising, and diagnostic applications, and enhances reproducibility and comparability in SCG research across the biomedical signal processing community.

Q: How were fiducial points validated?

Fiducial points were validated against gold-standard catheter-based measurements of aortic opening and closing events.

Validating the rigor and consistency of annotations, performed analyses for inter-annotator agreements resulted high agreements achieving strong correlations against gold-standard catheter-based measurements for aortic opening (AO) (r=0.926) and aortic closing (AC) (r=0.911).

Key Results

  • Correlation coefficients of r=0.926 for AO and r=0.911 for AC validate the accuracy of annotated fiducial points against catheter-based measurements.

  • Inter-annotator agreement for fiducial points was substantial, with Krippendorff’s alpha values of 0.762 for AO and 0.754 for MO.

Visual Evidence

Fig. 6 Unnormalized correlation plot of median annotator selected AO and AC compared against gold-standard AoP catheter  extracted AO and AC timings across all beats and pigs. Only heartbeats with an AoP DTW SQI score > 0.85 were used.

Fig. 6 Unnormalized correlation plot of median annotator selected AO and AC compared against gold-standard AoP catheter extracted AO and AC timings across all beats and pigs. Only heartbeats with an AoP DTW SQI score > 0.85 were used.

Clinical Snapshot

Evidence Rating

Relevance

high Priority

Confidence

Cornerstone

Relativity Score

5/5
Rigor
4/5
Novelty
5/5
Impact