Severity detection tool for patients with infectious disease.

Area of Science:

• Medical Informatics
• Biomedical Engineering
• Infectious Disease Research

Background:

• Hand, Foot, and Mouth Disease (HFMD) and tetanus are significant infectious diseases in low- and middle-income countries.
• Autonomic Nervous System Dysfunction (ANSD) is a primary cause of mortality in both HFMD and tetanus patients.
• Early detection of ANSD is crucial but challenging, hindering timely treatment and resource allocation.

Purpose of the Study:

• To develop and validate a machine learning approach for automated detection of ANSD levels in HFMD and tetanus patients.
• To utilize physiological data, specifically electrocardiogram (ECG) waveforms, collected via low-cost wearable sensors.
• To provide a proof-of-principle for improving diagnosis and treatment in resource-limited healthcare settings.

Main Methods:

• Extraction of efficient time and frequency domain features from ECG waveforms.
• Application of machine learning techniques to analyze extracted features for ANSD detection.
• Validation of the proposed method on independent datasets of HFMD and tetanus patients.

Main Results:

• The machine learning approach achieved encouraging performance in detecting ANSD.
• The extracted features demonstrated simplicity and generalizability.
• The proposed method outperformed standard heart rate variability analysis in accuracy.

Conclusions:

• Automated ANSD detection using machine learning and ECG data is feasible and effective.
• This approach can significantly aid in the early diagnosis and management of HFMD and tetanus.
• The technology holds potential to improve patient care and reduce mortality in low- and middle-income countries.