Contactless Vital Signs Monitoring in Telehealth: Platforms, Evidence, and Regulation

Contactless vital signs monitoring uses cameras, radar, or thermal sensors to measure heart rate, respiratory rate, and other physiological parameters without any wearable device or physical contact. In telehealth, this means a patient's vital signs can be captured during a routine video consultation using the webcam already built into their laptop or phone. The technology has moved from research curiosity to commercial deployment, driven partly by the pandemic and partly by steady improvements in the underlying algorithms.

What Contactless Vital Signs Monitoring Actually Means

The term covers several distinct technologies, each with different capabilities and maturity levels.

Remote photoplethysmography (rPPG) is the most common approach in telehealth. A standard RGB camera captures facial video, and software extracts the subtle skin color changes caused by blood pulsing through capillaries. From this signal, heart rate and respiratory rate can be derived. Some systems also attempt heart rate variability, oxygen saturation, and stress indicators. For a full technical explanation, see our rPPG algorithms deep dive.

Thermal imaging uses infrared cameras to detect heat patterns on the face and body. Breathing produces periodic temperature changes around the nostrils, enabling respiratory rate measurement. Fever screening via thermal cameras became ubiquitous during COVID-19, though accuracy for that specific use case was questionable. Thermal cameras are expensive and not standard consumer hardware, limiting telehealth applications.

Radar-based monitoring uses millimeter-wave or ultra-wideband radar to detect chest wall displacement from breathing and heartbeat. Radar works through clothing and in darkness, but requires dedicated hardware not present in consumer devices. It is more relevant for in-home monitoring stations than telehealth video calls.

For telehealth purposes, rPPG dominates because it works with existing consumer cameras. The rest of this article focuses primarily on camera-based approaches. Our contactless vital signs detection overview covers all modalities in more detail.

How COVID-19 Changed Everything

The pandemic did not invent contactless monitoring, but it created the conditions for rapid adoption. Three things happened simultaneously.

First, telehealth usage exploded. McKinsey estimated a 38x increase in telehealth utilization in the early months of the pandemic compared to pre-COVID baselines. Millions of patients who had never used video consultations were suddenly doing so regularly. Clinicians wanted vital signs from these visits. Patients did not always have home monitoring devices.

Second, infection control became paramount. In hospitals and clinics, any technology that reduced physical contact between healthcare workers and patients had immediate appeal. Contactless monitoring eliminated the need to attach pulse oximeters, blood pressure cuffs, and other sensors that required close proximity or shared equipment.

Third, funding poured in. Venture capital investment in digital health reached record levels in 2020 and 2021. Companies working on contactless vital signs attracted significant funding. Binah.ai raised $13.5 million in 2021. Nuralogix secured multiple funding rounds. Several other startups entered the space.

The pandemic also exposed a problem. When governments and employers deployed thermal cameras for fever screening at scale, independent studies showed the accuracy was poor. A 2021 study published in the BMJ found that thermal cameras missed a significant proportion of febrile individuals and flagged many who did not have fevers. This experience highlighted the gap between marketing claims and validated performance, a lesson that applies to all contactless monitoring technologies.

Current Platforms and Products

Several companies now offer contactless vital signs measurement for telehealth integration. Each takes a slightly different technical and commercial approach.

Binah.ai

Binah.ai provides an SDK that developers embed into telehealth and wellness applications. Their technology uses the smartphone or laptop camera to measure heart rate, respiratory rate, oxygen saturation, heart rate variability, and a stress index. The SDK processes video locally on the device, which addresses some privacy concerns since raw video does not need to be transmitted to a server.

Binah.ai has published validation studies and received CE marking for some of their measurements in Europe. They position themselves as a platform for other companies to build on rather than a consumer-facing product.

Nuralogix (Anura)

Nuralogix, a Toronto-based company, developed the Anura app and the Affex SDK. Their approach uses what they call "Transdermal Optical Imaging," which is essentially rPPG with additional signal processing. They claim measurement of heart rate, respiratory rate, blood pressure, and several wellness indicators from a 30-second selfie video.

The blood pressure claims are particularly notable because non-invasive blood pressure measurement from camera video remains scientifically controversial. Nuralogix has published peer-reviewed papers, but the clinical evidence for camera-based blood pressure accuracy at clinical-grade levels is still limited across the field as a whole.

Google Health Studies

Google has conducted extensive research on camera-based vital signs measurement. Their work includes using smartphone cameras to measure heart rate and respiratory rate, published in peer-reviewed venues. The Google Fit app briefly included heart rate and respiratory rate measurement features using the phone camera.

Google's approach benefits from massive computational resources and access to large, diverse datasets for training and validation. Their published research has advanced the state of the art, particularly in deep learning approaches to rPPG. However, Google has been cautious about clinical claims, positioning these features as wellness tools rather than medical devices.

Xperi (FaceMe Health)

Xperi, known for their media and entertainment technology, entered the health space with camera-based vital signs. Their technology targets integration into smart TVs, set-top boxes, and video conferencing platforms. The idea of measuring vital signs while watching TV or during a video call without any deliberate user action is appealing from a user experience perspective, though it raises distinct privacy questions.

Other Players

The field includes numerous other companies and research groups. CareAi, VeyeTal (acquired by Panasonic), and several Chinese companies (including platforms integrated into WeChat) offer camera-based vital signs measurement. Academic groups at universities including MIT, ETH Zurich, and the University of Oulu continue to push algorithmic boundaries.

How Integration into Telehealth Works

The typical integration follows a predictable pattern. Before or during a telehealth video call, the patient is prompted to hold still and look at their camera for 15 to 60 seconds. The contactless monitoring SDK, running either in the browser or a native app, captures facial video and processes it locally. Extracted vital signs are displayed to the patient and transmitted to the clinician's interface.

Some implementations run continuously during the video call, providing ongoing heart rate tracking. Others perform a discrete measurement at a specific point in the consultation.

The workflow integration matters as much as the technology. Clinicians need vital signs presented in familiar formats, ideally within their existing electronic health record (EHR) or telehealth platform. If the measurement requires a separate app, a different workflow, or manual data entry, adoption drops sharply. The most successful integrations are invisible to the clinician, with vital signs simply appearing in the patient's chart.

Latency is a practical concern. Real-time heart rate estimation requires at least 5 to 10 seconds of video, and reliable measurement typically needs 15 to 30 seconds. During a telehealth visit, this is easily managed. For continuous monitoring, algorithms must balance responsiveness against accuracy.

Clinical Validation: What the Evidence Shows

Published clinical evidence for contactless vital signs in telehealth settings is growing but still limited compared to established monitoring technologies. For context on the underlying technology validation, see our camera-based rPPG guide.

Heart Rate

Heart rate is the most validated contactless measurement. A systematic review by Huang et al. (2023) in Physiological Measurement examined 28 studies and found that rPPG heart rate measurement achieved a pooled mean absolute error of approximately 2 to 5 BPM under controlled conditions. Performance degraded with motion, poor lighting, and video compression.

A key study by Amelard et al. published in npj Digital Medicine (doi:10.1038/s41746-022-00606-z) evaluated camera-based vital signs measurement in a clinical population and found heart rate errors within clinically acceptable ranges for many patients, while identifying subgroups where performance was inadequate.

Respiratory Rate

Camera-based respiratory rate uses either chest motion tracking or modulation of the rPPG signal by breathing. Accuracy is generally good under controlled conditions, with errors of 1 to 3 breaths per minute. However, shallow breathing and speech during a telehealth call can confound measurements.

Blood Pressure

Camera-based blood pressure estimation remains the most controversial claim. The physiological basis for extracting blood pressure from facial video is weak. Blood pressure is determined by cardiac output and peripheral vascular resistance, neither of which is directly observable from skin surface optical signals.

Some studies report correlations between camera-derived features and blood pressure, but correlation is not the same as clinical-grade measurement accuracy. A study by Schrumpf et al. (2021) in Sensors (doi:10.3390/s21124016) provided a systematic assessment of deep learning methods for camera-based blood pressure estimation, highlighting the significant challenges that remain.

The IEEE and other standards bodies have called for more rigorous validation of contactless blood pressure claims. Until large-scale, multi-site clinical trials demonstrate consistent accuracy across diverse populations, blood pressure from facial video should not be used for clinical decisions.

Oxygen Saturation (SpO2)

Estimating SpO2 from consumer cameras is technically possible in principle, since the ratio of light absorption at different wavelengths changes with blood oxygen levels. In practice, consumer RGB cameras are poorly suited for this because their spectral response curves are broad and overlapping, making it difficult to isolate the specific wavelength-dependent absorption changes that pulse oximeters exploit.

Some companies claim SpO2 measurement from smartphone cameras. The evidence for clinically useful accuracy is thin. Dedicated pulse oximeters use carefully calibrated LEDs at specific wavelengths (typically 660 nm and 940 nm). A phone camera with broadband ambient light is a poor substitute.

FDA and CE Regulatory Pathways

Contactless vital signs software is classified as Software as a Medical Device (SaMD) under most regulatory frameworks. The regulatory pathway depends on the intended use and risk classification. For more detail on FDA pathways specifically, see our rPPG FDA regulatory status guide.

FDA (United States)

The FDA uses the International Medical Device Regulators Forum (IMDRF) framework to classify SaMD based on the seriousness of the health condition and the significance of the information to the clinical decision.

A wellness heart rate measurement (not intended for diagnosis or treatment decisions) falls into the lowest risk category and may not require FDA clearance at all. A heart rate monitor intended for clinical use during telehealth consultations, where the reading could influence treatment, requires at least a 510(k) clearance and possibly a De Novo pathway if no suitable predicate device exists.

The FDA has been increasingly active in the digital health space. Their Digital Health Center of Excellence provides guidance on SaMD classification and premarket requirements. For contactless monitoring, the FDA expects clinical validation studies demonstrating accuracy across diverse populations, including different skin tones, ages, and clinical conditions.

As of early 2026, no purely camera-based contactless vital signs system has received full FDA clearance for clinical diagnostic use. Several have 510(k) clearance for wellness or fitness purposes.

CE Marking (European Union)

Under the EU Medical Device Regulation (MDR 2017/745), contactless vital signs software intended for medical purposes is classified as a medical device. Classification depends on the intended purpose, with most clinical vital signs measurements falling into Class IIa or IIb.

CE marking requires conformity assessment by a Notified Body, including review of clinical evidence, risk management, and quality management systems. Some contactless monitoring companies have achieved CE marking for specific measurements and intended uses in Europe ahead of FDA clearance in the US.

The Regulatory Gap

There is a gap between what companies market and what regulators have cleared. Many contactless vital signs apps and SDKs are available commercially with health-related claims, but without medical device clearance. They operate in a gray zone, often positioning measurements as "wellness" rather than "clinical" to avoid regulatory requirements.

This matters because patients and clinicians may not understand the distinction. A heart rate reading displayed during a telehealth visit looks the same whether it comes from a cleared medical device or an uncleared wellness tool. The accuracy and reliability may be very different.

Privacy and Data Security

Camera-based vital signs measurement involves processing facial video. This raises legitimate privacy concerns that go beyond standard telehealth data handling. We explore the broader ethical dimensions in our rPPG privacy and data ethics analysis.

Data minimization: The best implementations process video locally on the patient's device and transmit only the extracted vital signs, not the raw video. This is both a privacy feature and a bandwidth optimization. If raw facial video is transmitted to a server for processing, it falls under biometric data regulations in many jurisdictions.

HIPAA considerations (US): Vital signs data generated during a telehealth visit is protected health information (PHI) under HIPAA. The contactless monitoring vendor must be a Business Associate with appropriate agreements in place. If the processing occurs in a third-party cloud, that cloud provider is also subject to HIPAA requirements.

GDPR considerations (EU): Facial video is biometric data under GDPR. Processing requires explicit consent and a lawful basis. The right to erasure applies to any stored video or derived biometric data. Data Protection Impact Assessments are likely required for telehealth deployments using facial analysis.

Informed consent: Patients should understand that their camera is being used to measure vital signs, how the data is processed, where it is stored, and who has access. This consent should be separate from general telehealth consent. Many current implementations fall short on transparency.

Secondary use risks: Facial video captured for vital signs measurement could theoretically be used for other purposes, such as emotion detection, identity verification, or behavioral analysis. Strong data governance policies and technical controls are needed to prevent scope creep.

Barriers to Clinical Adoption

Despite the technology's promise and commercial availability, clinical adoption in telehealth remains limited. Several barriers explain why.

Accuracy confidence: Clinicians are trained to trust validated instruments. A blood pressure reading from an inflatable cuff has decades of clinical evidence behind it. A heart rate reading from a webcam does not yet carry the same credibility. This skepticism is healthy.

Workflow integration: Adding a vital signs measurement step to a telehealth visit requires software integration, clinician training, and patient instructions. In a 15-minute video consultation, even 30 seconds of measurement time feels significant if it disrupts the conversational flow.

Liability: If a clinician makes a treatment decision based on a contactless vital signs reading that turns out to be inaccurate, who bears liability? The clinician? The telehealth platform? The SDK vendor? This question is largely untested in court, and the ambiguity makes risk-averse health systems cautious.

Reimbursement: In many healthcare systems, vital signs measurement during a telehealth visit is not separately reimbursed. If there is no financial incentive to capture vital signs and the technology has uncertain accuracy, the practical motivation for adoption is low.

Digital divide: Contactless monitoring requires a reasonable camera and adequate lighting. Patients using older devices, poor internet connections, or dim rooms will get unreliable measurements. This risks widening health disparities rather than reducing them.

The Path Forward

Several developments could accelerate clinical adoption over the next few years.

Better algorithms: Deep learning models continue to improve, particularly for handling motion, diverse skin tones, and low-quality video. Self-supervised learning approaches reduce the need for expensive labeled clinical datasets. For a review of current algorithmic approaches, see our rPPG guide.

Standardized benchmarks: The field needs agreed-upon benchmark datasets and evaluation protocols. Currently, every company validates on different populations with different reference devices, making comparison nearly impossible. Efforts like the rPPG benchmark from the IEEE are steps in the right direction.

Regulatory clarity: As the FDA and EU regulators develop more specific guidance for contactless monitoring SaMD, the pathway to clearance becomes more predictable. This encourages investment in clinical trials.

Integration with remote patient monitoring: Contactless vital signs measurement during telehealth visits is one use case. Continuous or daily measurement at home, using a tablet or smart display, is another. Combining both into a coherent remote patient monitoring program adds more clinical value than either alone. For more on home monitoring applications, see our PPG home telehealth monitoring guide.

Clinical champions: Adoption of new clinical technology almost always requires physician champions who advocate based on personal experience. Early adopter clinicians who integrate contactless monitoring into their telehealth practice and publish their results will drive broader acceptance.

The technology works well enough today for basic heart rate and respiratory rate monitoring under reasonable conditions. It does not yet replace dedicated monitoring devices for clinical decision-making. The honest position is somewhere between the hype (contactless monitoring will replace all wearables) and the skepticism (it will never be accurate enough). Steady progress in algorithms, validation, and regulation is closing the gap.

Frequently Asked Questions

Can vital signs really be measured through a video call?

Yes, for heart rate and respiratory rate. The technology uses subtle color changes in facial skin caused by blood flow to extract heart rate, and chest or facial motion to extract respiratory rate. Accuracy is typically within 2 to 5 BPM for heart rate under good conditions (stable lighting, minimal movement, decent camera quality). Blood pressure and SpO2 claims from video are less well supported by evidence.

Do I need special equipment for contactless vital signs in telehealth?

No special hardware is needed for basic rPPG-based measurement. A standard laptop webcam or smartphone front camera is sufficient. Better cameras (higher resolution, higher frame rate, less compression) improve accuracy. Good, stable lighting helps significantly. You do not need an infrared camera or any additional sensors for the most common use case.

How long does a contactless measurement take?

Most systems require 15 to 60 seconds of the patient sitting relatively still and facing the camera. Some continuous monitoring approaches extract ongoing measurements throughout the telehealth call, but the first reliable reading still takes at least 10 to 15 seconds of usable video.

Is contactless vital signs monitoring FDA-approved?

As of early 2026, no camera-based contactless vital signs system has full FDA clearance for clinical diagnostic use. Some products have FDA clearance for wellness or fitness applications, which is a lower regulatory bar. Several companies have CE marking in Europe for specific measurements. The regulatory pathway is active, and clearances may come as clinical validation data accumulates.

Is my privacy protected during contactless measurement?

It depends on the implementation. The best systems process facial video entirely on your device and never transmit raw video to a server. Only the extracted vital signs numbers are sent. You should ask your telehealth provider whether video is processed locally or in the cloud, and whether any facial video is stored. Contactless monitoring data falls under HIPAA (US) and GDPR (EU) protections when used in healthcare contexts.

Can contactless monitoring detect COVID-19 or other infections?

Not directly. Contactless monitoring can measure vital signs like heart rate, respiratory rate, and (with limited accuracy) temperature via thermal cameras. Abnormal vital signs may indicate illness, but they cannot identify a specific infection. Elevated heart rate and respiratory rate are nonspecific findings that occur with many conditions. Fever screening via thermal cameras during the pandemic had mixed accuracy in real-world settings.

Will contactless monitoring replace wearable health devices?

Unlikely in the near term. Wearable devices like smartwatches and pulse oximeters have direct skin contact, dedicated sensors, and controlled optical paths, giving them inherent accuracy advantages. Contactless monitoring fills a different niche: situations where wearables are unavailable, impractical, or undesirable. The two approaches are complementary. A patient might use a smartwatch for daily tracking and contactless measurement during telehealth visits when they do not have a dedicated device handy.