newsnews.ai

Smartphone Camera Can Passively Monitor Heart Rate During Normal Use

Researchers have developed a system that uses a phone's front-facing camera to detect blood flow changes in the face and track heart rate without wearable devices.

By NewsNews AI
shallow focus photo of man in gray collared top taking selfie
shallow focus photo of man in gray collared top taking selfie·Photo: Hc Digital on Unsplashunsplash

Passive Monitoring Technology

Researchers have developed a system called Passive Heart-Rate Monitoring (PHRM) that allows smartphones to automatically track a user's heart health during normal device operation. Unlike traditional methods, this technology does not require the use of a smartwatch, fitness tracker, or a deliberate manual check by the user.

The PHRM system utilizes the smartphone's front-facing camera to detect subtle changes in blood flow across the user's face. The system is designed to run passively in the background and automatically initiate video capture via the camera to monitor heart rate (HR) during everyday life.

Machine Learning and Data Training

To enable the system to recognize heart rate patterns, the research team employed a large-scale training process. The model was first trained using 192,353 video recordings obtained from 485 different people. Following this initial phase, the researchers tested the system's efficacy using a separate dataset consisting of 162,546 recordings collected from 211 individuals.

The model utilizes a distributional output to express uncertainty in its estimations. According to the research, if there is a high degree of uncertainty—such as during periods of extreme motion—the probability distribution flattens. This approach differs from standard regression models, which are forced to provide a single point estimate that may be erroneous in the presence of noise or movement.

Calculating Daily Resting Heart Rate

Beyond instantaneous heart rate detection, the researchers developed an algorithm to derive a user's daily resting heart rate (RHR). This is achieved by aggregating HR predictions throughout the day.

The aggregation process utilizes both the confidence of individual predictions and a Kalman filter to ensure accuracy. By combining these elements, the PHRM system can provide a more stable estimate of the user's resting heart rate over time despite the variability of passive data collection.

Sources (7)Open

Topics

How NewsNews AI made this storyOpen

NewsNews AI researched this story across 7 sources, drafted it, and ran the result through an independent editorial pass. It cleared editorial review on first pass.

  • 7 sources cited · linked in full at the bottom of the article
  • Image license verified · unsplash
  • Independent editorial pass · approved

From the editor

All factual claims in the body and key facts were verified against the provided source snippets. Source [^4] (News-Medical) supports the PHRM description, training/testing dataset figures, and the passive monitoring framing. Source [^2] (Nature) supports the distributional output, uncertainty handling, Kalman filter, and daily RHR aggregation details. No fabricated quotes, no contradictions, no unsupported claims, and no single-source saturation issues were found. The headline and dek accurately reflect the article content.

More about our editorial process

Feedback

We want to hear from you, especially when something is wrong. No signup, no email required.

Keep reading