Wearables Detect Malaria Early in a Controlled Human-Infection Study

Area of Science:

• Biomedical Engineering
• Infectious Disease Research
• Wearable Technology

Background:

• Controlled human infection studies offer rigorous evaluation of medical interventions.
• Wearable devices can monitor physiological signals for health insights.
• Early infection detection is crucial for timely treatment and public health.

Purpose of the Study:

• To assess the feasibility of using a commercial smartwatch for infection detection.
• To develop and validate an algorithm for predicting infection using wearable data.
• To evaluate the performance of the 2B-Healthy algorithm in a controlled human malaria infection (CHMI) model.

Main Methods:

• Ten subjects underwent CHMI, wearing a smartwatch to collect heart rate, skin temperature, and acceleration data.
• A Bayesian-based algorithm, 2B-Healthy, was developed to estimate infection probability.
• Eight control subjects provided data to determine the algorithm's false-positive rate.

Main Results:

• 2B-Healthy achieved 78% sensitivity in detecting malaria infection.
• The algorithm identified infection an average of 6 days before parasitemia in six subjects.
• A low false-positive rate of 6% per week was observed in control subjects.

Conclusions:

• The 2B-Healthy algorithm reliably detected infection prior to symptom onset using smartwatch data.
• Wearable devices show feasibility as a screening tool for early infection warning.
• Further research can establish 2B-Healthy as a basis for a wearable infection-detection platform.