Can a Doctor See My Child's Breathing Rate Without Touching Them During a Video Call?

A parent on a pediatric video call is usually doing something the camera cannot replicate: watching the chest rise and fall, counting breaths, listening for the catch that means a cough is turning into something worse. Respiratory rate is one of the earliest and most reliable warning signs in children, yet for most of the history of telemedicine it has been invisible to the provider on the other end of the screen. The question parents increasingly ask their telehealth provider, and the question telehealth software vendors increasingly hear from clinical buyers, is whether a doctor can actually see a child's breathing rate without touching them. Contactless vitals telemedicine, built on camera-based signal extraction, is the technology trying to answer yes.

In adult populations, remote photoplethysmography has reported respiratory rate errors as low as 0.40 plus or minus 0.11 breaths per minute, while a recent pediatric study found only weak correlation (Rs = -0.02) for respiratory rate in children aged 5 to 16, a gap that defines the current engineering frontier.

How contactless vitals telemedicine estimates breathing rate

The short answer for parents is that a doctor cannot literally feel a child breathe over video, but software can estimate the breathing rate from the video signal itself. Contactless vitals telemedicine relies primarily on remote photoplethysmography, or rPPG, a method that reads tiny color changes in the skin caused by blood flowing with each heartbeat. Respiratory rate is derived indirectly, because breathing modulates the heartbeat signal and also produces small, rhythmic movements of the chest, shoulders, and head that a camera can track frame by frame.

There are two broad technical routes a telehealth platform can take, and they behave very differently with a wriggling toddler than with a still adult.

• Signal-based extraction (rPPG): the algorithm isolates the pulse waveform from facial skin pixels, then infers respiratory rate from the respiratory sinus arrhythmia modulation of that waveform.
• Motion-based extraction: the algorithm tracks periodic chest, shoulder, or head movement directly, often using Eulerian video magnification or optical flow, which can work even when the face is partly turned away.

For a child who will not sit still, motion-based methods and hybrid approaches often carry more weight, because the respiratory signal embedded in the pulse waveform degrades quickly with movement. This is the central reason pediatric breathing rate is harder than adult heart rate.

| Approach | Primary signal | Best pediatric fit | Movement tolerance | Maturity for breathing rate | | --- | --- | --- | --- | --- | | rPPG facial signal | Skin color change | Older children, cooperative teens | Low | Validated for HR, refining for RR | | Chest or shoulder motion | Periodic body movement | Infants, sleeping children | Moderate | Promising in NICU and sleep studies | | Head movement tracking | Micro head oscillation | Neonates, still infants | Moderate | Emerging, RGB camera based | | Depth or time-of-flight camera | 3D chest displacement | Ages 6 months to 12 years | Higher | Prototype, accurate in trials |

What this means for a pediatric video visit

When a telehealth platform integrates contactless vitals, the breathing rate estimate appears as a derived number, usually presented to the provider with a confidence indicator rather than a single hard value. For parents, the practical reality is straightforward: the better the lighting, the steadier the child, and the closer the framing of the upper body, the more usable the estimate. None of this requires a cuff, a clip, or a wearable on the child.

The value is not that contactless measurement replaces a clinical exam. It is that it gives a remote provider a structured data point where previously there was only a parent's verbal description. A breathing rate trend across a visit, or across several visits, can change a triage decision.

• No hardware to attach to a child who resists clips and bands.
• No fragile-skin concerns associated with adhesive sensors in infants.
• Shorter visit setup time, since the camera is already running.
• A measurement the provider can document rather than estimate by eye.

Industry applications for telehealth software vendors

For telemedicine software vendors evaluating contactless vitals telemedicine for pediatric care, the use cases fall into distinct product categories, each with its own integration and accuracy expectations.

Acute pediatric triage

The highest-value scenario is the sick-child visit, where elevated respiratory rate is a recognized red flag for respiratory distress, dehydration, and sepsis. A platform that surfaces a breathing rate estimate during a video triage call gives the provider a quantitative anchor for the escalation decision. Because acute visits often involve a distressed, moving child, vendors should design for confidence-scored output and graceful degradation rather than a false promise of a single precise number.

Overnight and at-home monitoring

Motion-based contactless monitoring is well suited to sleeping children, where the body is relatively still and the upper body is in frame. Vendors building remote patient monitoring features can use this to capture respiratory trends across the night without waking the child, which connects directly to the chronic and post-acute monitoring market.

Chronic respiratory management

For children with asthma or other chronic respiratory conditions, longitudinal breathing rate data captured during routine video check-ins supports trend analysis. The clinical value here is comparison over time rather than a single instantaneous reading, which is more forgiving of per-measurement variability.

Current research and evidence

The evidence base is honest about both promise and limitation, and telehealth vendors should be equally honest with clinical buyers. A two-phased pediatric study reported that rPPG was feasible and acceptable to children and their caregivers across a range of ages, and that heart rate correlation was good in older children aged 12 to 16. The same line of research found that respiratory rate and oxygen saturation algorithms still require refinement across all pediatric age groups, with one study reporting weak respiratory rate correlation (Rs = -0.02) in children aged 5 to 16. Movement during measurement is repeatedly named as the dominant source of error.

By contrast, adult rPPG respiratory rate performance is considerably more mature, with reported absolute mean errors near 0.40 breaths per minute and accuracy figures up to 96 percent in some studies. The gap between adult and pediatric performance is not a flaw in the physics; it is a consequence of motion, smaller anatomy, and variable cooperation.

Newer approaches are narrowing that gap. A 2024 study described an RGB video method using Eulerian video magnification to estimate respiratory rate and detect breathing absence in infants in the NICU, performing under challenging conditions. Separate work on head-movement-based monitoring with a standard RGB camera demonstrated respiratory rate estimation and apnea-style breathing-absence detection. An AI-based prototype using a time-of-flight depth camera reported reliable, non-invasive respiratory monitoring in children from 6 months to 12 years, including detection of abnormal breathing patterns. The prospective rMonitoped1 trial is enrolling 600 pediatric participants to compare rPPG heart rate and respiratory rate against standard clinical monitoring, with results expected in 2026, which should give vendors a much firmer evidence base.

The future of contactless pediatric vitals

The direction of travel is toward hybrid pipelines that fuse the pulse-derived respiratory signal with direct motion tracking, then weight them by a real-time quality score. For pediatric care specifically, this matters more than any single algorithm, because a child's cooperation is unpredictable and a robust system has to know when to trust its own output. Depth-camera and standard-webcam approaches will likely converge in software, with platforms selecting the best available signal from whatever the device offers.

For telehealth software vendors, the strategic implication is that contactless vitals telemedicine is moving from a single-vital novelty to an expected layer of the pediatric visit. The platforms that win will be the ones that present confidence transparently, document defensibly, and integrate vitals capture into the existing video workflow rather than bolting on a separate measurement step. Parents will not adopt a feature that asks them to wrestle a child into a perfect pose. Providers will not trust a number without context.

Frequently asked questions

Can a doctor really measure my child's breathing rate over a video call?

A doctor cannot physically touch the child, but contactless vitals software can estimate breathing rate from the video by reading subtle skin color changes and chest or head movement. It is best treated as a supportive data point alongside the clinical exam, not a standalone diagnosis, and accuracy improves with good lighting and a steady child.

Is contactless breathing rate measurement accurate for children?

Accuracy is strong in adults but still maturing in children. Research shows good heart rate correlation in older children while respiratory rate algorithms are being refined, with movement being the main challenge. Sleeping infants and cooperative older children currently yield more reliable estimates than active toddlers.

Does my child need to wear any device for contactless vitals?

No. The defining feature of contactless vitals telemedicine is that nothing is attached to the child. The estimate is derived entirely from the existing camera feed, which removes the adhesive, clip, and band issues that make traditional pediatric monitoring difficult.

How should a telehealth platform present these vitals to providers?

Best practice is to display the estimate with a confidence or signal-quality indicator, support trend views across a visit or multiple visits, and avoid presenting a single number as a hard clinical reading. This keeps clinical interpretation in the provider's hands while still adding documentable value.

Circadify is working on this exact problem space, building an rPPG SDK that adds real-time contactless vitals to telehealth platforms without any patient hardware, designed to fit into existing video visit workflows including pediatric care. Telehealth software teams evaluating contactless vitals for their product can review the platform demo and SDK documentation at circadify.com/custom-builds.