Event Cameras: A New Way to Measure Heart Rates
Research explores the use of event cameras for non-contact heart rate monitoring.
― 6 min read
Table of Contents
- How Event Cameras Work
- Why Measure Heart Rate Noninvasively?
- Understanding Periodicity in Heart Rate
- Using Event Cameras for Heart Rate Measurement
- Analyzing the Data
- Results of the Experiments
- Advantages of Event Cameras
- Challenges and Considerations
- Future Research Directions
- Conclusion
- Original Source
- Reference Links
Event Cameras, also called neuromorphic cameras, are a new type of technology that provide some key benefits over regular cameras. They can capture images very quickly and use less power while focusing only on what is important in a scene.
This research investigates whether event cameras can be used to continuously monitor Heart Rates without needing to touch the skin. By using an event camera to look at the tiny changes in the skin where blood flows, we tested if it could effectively measure heart rates. We collected video data from 25 Participants of different ages and skin tones, and we compared the heart rates detected by the camera against traditional methods.
The results showed that using event cameras to detect heart rates can work. However, we also noted some challenges, like flickering light and natural hand movements during the data collection.
How Event Cameras Work
Event cameras are different from traditional cameras. Regular cameras capture images at fixed intervals, usually 25 or 30 times per second. This can lead to problems when trying to catch very fast movements, as some information might be missed.
In contrast, event cameras only respond to changes in light. They record when there are significant changes in brightness, allowing them to capture fast movements without losing details. This means they can produce thousands of images a second, allowing for clearer and more accurate Recordings of quick events.
These cameras have a wide range of applications, especially in areas requiring fast responses, like robotics or wearables, where power efficiency and adaptability to different Lighting Conditions are crucial.
Why Measure Heart Rate Noninvasively?
Measuring heart rate is important for several health-related applications. Traditional methods require physical contact, like placing sensors on the skin. These methods work well, but they can be inconvenient.
Using non-contact methods, such as event cameras, could lead to more comfortable and easier monitoring of heart rates, especially for people who need to track their heart rates continuously or in different situations, like while driving or exercising.
Understanding Periodicity in Heart Rate
Heart rate has a consistent pattern known as periodicity, where the heart beats at regular intervals. For example, when resting, a person's heart might beat about 70 times in a minute. When active, this number can increase gradually.
Detecting this regular pattern is critical for understanding heart rate. Any irregularity in the pattern can indicate potential health issues. Thus, measuring periodicity can provide insights into a person's health.
Traditional heart rate monitors, like those found in smartwatches or heart rate sensors, often use light sensors to capture blood flow through the skin. The light detects variations that correspond to heartbeats, allowing these devices to be effective in measuring heart rates.
Using Event Cameras for Heart Rate Measurement
To determine if event cameras could effectively measure heart rates, we conducted experiments where we had participants wear Apple watches to record their actual heart rates. We also marked a dot on their wrist, where we wanted the camera to focus. This dot would help the event camera see any movement caused by blood flow beneath the skin.
Participants relaxed and held their arms still while the camera recorded for around 12 to 15 seconds. After capturing this initial data, we had them perform light exercises to elevate their heart rates. We repeated the recording process during this time.
The data collected included a range of heart rates from different subjects with various skin tones and age groups. Each participant contributed multiple recordings, allowing us to gather a robust dataset for analysis.
Analyzing the Data
After collecting all the recordings, we analyzed the event camera data. The camera generated a stream of information based on brightness changes, which we examined to determine heart rates.
To process the data, we looked at regions where the pixel activity was highest. Each area was broken down into smaller parts, allowing us to find the prevalent heart rate frequencies in those regions.
Using different algorithms, we calculated the heart rates and compared these estimates against the actual heart rates recorded by the Apple watches. This way, we could evaluate how accurate the event camera measurements were.
Results of the Experiments
Out of 46 recordings from 25 participants, we successfully detected heart rates in 40 cases. In three instances, natural hand movements hindered detection, while for another three, we could not pinpoint the cause of nondetection.
The results showed that our heart rate detection method using event cameras was effective. The highest difference between the actual heart rate and the detected rate was just 5 beats per minute, with most estimates being accurate within one beat.
The average error in our estimates was lower than 2 beats per minute for resting and elevated heart rates. This performance compares favorably with other non-contact heart rate measurement methods.
Advantages of Event Cameras
Event cameras bring various advantages over traditional methods. They can work well in different lighting conditions, handle fast motions, and provide real-time results.
Their high sensitivity and precise recording capabilities make them excellent for applications needing continuous monitoring without physical contact. This characteristic is particularly useful for fields like healthcare and automotive safety.
Challenges and Considerations
Despite the promising results, some challenges remain. Participants needed to stay as still as possible during recordings, which could be difficult due to natural body movements. Additionally, variations in lighting could affect the quality of the recordings.
To improve the accuracy and reliability of heart rate detection, future research should focus on optimizing the algorithm for real-time applications, accommodating for natural movements, and ensuring consistent lighting conditions.
Future Research Directions
This study opens avenues for further exploration. Developing applications that use event cameras for real-time heart rate detection could enhance monitoring in various settings, such as hospitals or fitness centers.
Researchers should also focus on making sure these technologies can handle changing lighting conditions and different body motions seamlessly. Handling privacy and ethical considerations will be crucial as these technologies become more widely used.
Conducting real-world tests will help refine the technology and ensure it meets user needs effectively. Understanding how comfortable and easy it is for people to use these systems will be key to their success.
Conclusion
In conclusion, using event cameras for measuring heart rates is a feasible and promising idea. By accurately capturing changes in skin movement related to blood flow, event cameras can serve as effective tools for non-contact heart rate monitoring.
This research lays the groundwork for innovative applications in health monitoring, driver safety, and fitness tracking, enhancing how we understand and manage heart health in various environments. Future studies will be essential to refine these methods and explore their full potential in real-world scenarios.
Title: Heart Rate Detection Using an Event Camera
Abstract: Event cameras, also known as neuromorphic cameras, are an emerging technology that offer advantages over traditional shutter and frame-based cameras, including high temporal resolution, low power consumption, and selective data acquisition. In this study, we propose to harnesses the capabilities of event-based cameras to capture subtle changes in the surface of the skin caused by the pulsatile flow of blood in the wrist region. We investigate whether an event camera could be used for continuous noninvasive monitoring of heart rate (HR). Event camera video data from 25 participants, comprising varying age groups and skin colours, was collected and analysed. Ground-truth HR measurements obtained using conventional methods were used to evaluate of the accuracy of automatic detection of HR from event camera data. Our experimental results and comparison to the performance of other non-contact HR measurement methods demonstrate the feasibility of using event cameras for pulse detection. We also acknowledge the challenges and limitations of our method, such as light-induced flickering and the sub-conscious but naturally-occurring tremors of an individual during data capture.
Authors: Aniket Jagtap, RamaKrishna Venkatesh Saripalli, Joe Lemley, Waseem Shariff, Alan F. Smeaton
Last Update: 2023-09-21 00:00:00
Language: English
Source URL: https://arxiv.org/abs/2309.11891
Source PDF: https://arxiv.org/pdf/2309.11891
Licence: https://creativecommons.org/licenses/by/4.0/
Changes: This summary was created with assistance from AI and may have inaccuracies. For accurate information, please refer to the original source documents linked here.
Thank you to arxiv for use of its open access interoperability.