Back to Evidence Hub
Case Study S045
2025 Release

Parlato 2025: FOSTER Multimodal Dataset

A. Parlato et al.
Access Paper

Quick Conclusion: S045 (FOSTER) is a high-quality 2025 dataset that represents the future of multimodal 'Force-based' monitoring. For OpenSCG, it provides the precise temporal alignment needed to train models that combine vibrations (SCG) with acoustic data (PCG), enabling more robust detection of heart valve events.


📊 Key Accuracy Metrics

MetricResult
Total Data7 minutes per subject @ 10 kHz sampling rate
SCG SNR83.73 ± 0.53 dB
IBI Accuracy (dHF-FCG vs ECG)R2 > 0.99, Bias 0.0 ms, LoA ~±6 ms
Repositoryosf.io/3u6yb


🔍 Study Analysis

Objective & Population

Dataset Description / Technical Validation Study. Cohort: 40 healthy subjects (20 males, 20 females), Age 26.93 ± 7.09 (N=40).

What it Supports

The study supports the integration of multiple mechanical sensing modalities (SCG, FCG, PCG) for a holistic view of heart function. It proves that mechanical sensors can track heart and breathing rates with high precision (unit slopes vs ECG) and provides a unique high-frequency (10 kHz) dataset for developing advanced AI algorithms.

What it Does Not Support

This study does not provide diagnostic evidence for heart failure or other pathologies, as the cohort consisted of young, healthy volunteers.


🛠 Technical Context

Featured Illustration

Fig. 8  Histograms of the errors on the IBI estimates between the reference ECG signal and the FCG signal  components: (a) histogram of errors on IBIs between ECG and dHF-FCGPVDF; (b) histogram of errors on IBIs  between ECG and dHF-FCGPZT; (c) histogram of errors on IBIs between ECG and HS-FCGPVDF; (d) histogram  of errors on IBIs between ECG and HS-FCGPZT.

Fig. 8  Histograms of the errors on the IBI estimates between the reference ECG signal and the FCG signal components: (a) histogram of errors on IBIs between ECG and dHF-FCGPVDF; (b) histogram of errors on IBIs between ECG and dHF-FCGPZT; (c) histogram of errors on IBIs between ECG and HS-FCGPVDF; (d) histogram of errors on IBIs between ECG and HS-FCGPZT.

Study Snapshot

Metadata Summary

Target Population

40 healthy subjects (20 males, 20 females), Age 26.93 ± 7.09

N

Sample Size

40 Subjects

Validated Metric

7 minutes per subject @ 10 kHz sampling rate

Critical Appraisal
supporting

Validated a high-resolution multimodal dataset for synchronized cardio-respiratory mechanical sensing.