Classification of Decompensated Heart Failure From Clinical and Home Ballistocardiography

Area of Science:

• Biomedical Engineering
• Cardiovascular Health
• Remote Patient Monitoring

Background:

• Heart failure (HF) management requires continuous patient monitoring to prevent adverse events.
• Traditional monitoring methods can be resource-intensive, leading to hospital readmissions.

Purpose of the Study:

• To develop a non-invasive method for home monitoring of heart failure patients.
• To analyze ballistocardiogram (BCG) signals for assessing patient clinical status and reducing hospitalizations.

Main Methods:

• Collected high-quality BCG signals from HF patients at home post-discharge.
• Preprocessed BCG recordings to remove artifacts and extract cardiovascular parameters.
• Utilized extracted features for classifying HF patient status from BCG data.

Main Results:

• Achieved an Area Under the Curve (AUC) score of 0.78 for HF status classification.
• Demonstrated the feasibility of collecting high-quality BCG signals in a home setting.

Conclusions:

• High-quality BCG signals can be acquired remotely to monitor HF patients' clinical status.
• This approach shows potential for improving outpatient HF management and reducing healthcare utilization.