Is There a Way My Online Physical Therapist Can See My Circulation While I Exercise?

A patient mid-squat on a video call presents a problem that standard telehealth was never built to solve. The physical therapist on the other end can see the movement, count the repetitions, and correct the form. What they cannot see is what is happening underneath: whether the heart rate is climbing into a safe training zone, whether circulation is keeping pace with the effort, or whether the patient is quietly overexerting. This gap is exactly where telehealth vital signs integration is becoming a defining feature for rehabilitation platforms, and it is the reason camera-based monitoring during active exercise has moved from a research curiosity to a real product requirement for software vendors serving physical therapy and cardiac rehab markets.

A 2024 study of real-time online cardiac telerehabilitation using wearable sensors found the approach as effective as gym-based exercise programs for post-myocardial-infarction patients, with comparable improvements in daily steps and walking distance, reinforcing that remote exercise monitoring can match in-person standards when physiological data is captured continuously.

Why telehealth vital signs integration matters during movement, not just at rest

Most telehealth platforms capture vitals the easy way: the patient sits still, looks at the camera for thirty seconds, and a heart rate appears. Rehabilitation breaks that model. The clinical value in physical therapy and cardiac rehab is in the data during exertion, not before it. A therapist guiding a recovery program needs to know how circulation responds to load, how quickly the heart rate recovers between sets, and whether the patient is training inside the prescribed range.

Remote photoplethysmography, or rPPG, is the technology that makes this possible without strapping hardware to the patient. It reads tiny color changes in facial skin caused by blood volume pulsing through capillaries, using only the standard webcam already running the video visit. The hard part has never been the resting measurement. It is movement. Motion introduces noise that historically wrecked contactless heart rate estimates, which is why telehealth vital signs integration for active exercise is a more demanding engineering problem than the seated check-up most platforms ship today.

The progress here is recent and measurable. Research published in 2023 on enhanced contactless heart rate monitoring with motion artifact removal reported accuracy as tight as 2.29 beats per minute during activities such as yoga by analyzing head motion to discard unreliable signal windows. A separate two-stage motion artifact reduction method for facial video achieved a root mean square error of 3.41 beats per minute. These are not laboratory-only numbers anymore; they are the kind of tolerances that make exercise-time monitoring clinically usable.

Comparing monitoring approaches for online rehabilitation

Rehabilitation platforms evaluating how to capture circulation during exercise generally weigh three paths. Each carries different cost, friction, and accuracy tradeoffs.

| Monitoring approach | Patient hardware | Works during active motion | Setup friction | Best fit | | --- | --- | --- | --- | --- | | Chest-strap ECG / wearable biosensor | Required, must be charged and paired | High accuracy under motion | High; shipping, onboarding, support | Cardiac rehab with strict telemetry needs | | Fingertip pulse oximeter | Required, low cost | Poor; movement disrupts contact | Medium | Brief spot checks at rest | | Contactless rPPG via webcam | None | Improving rapidly with motion-tolerant algorithms | Minimal; uses existing video feed | Telerehab, PT, general exercise monitoring | | Manual patient self-report | None | Not real-time | Low | Adherence and pain logging only |

The strategic pull toward rPPG is straightforward. Every other physiological method asks the patient to acquire, maintain, and correctly wear a device, and adherence drops with every added step. A camera-based pipeline removes that barrier entirely, which matters most for the older and post-acute populations who dominate rehabilitation caseloads.

Key considerations when assessing exercise-time vitals capture:

• Signal quality gating: the system should flag and suppress low-confidence readings during heavy motion rather than report a misleading number.
• Continuous trends over single values: recovery slope and heart-rate variability across a session carry more rehab value than one peak figure.
• Latency: a therapist correcting effort in real time needs sub-second feedback, not a post-session report.
• Privacy posture: video-derived vitals fall under the same protected health information rules as any other clinical data.

Industry applications for circulation monitoring in telerehab

Physical therapy and remote therapeutic monitoring

In 2024, the Centers for Medicare and Medicaid Services expanded how physical and occupational therapists can bill for remote therapeutic monitoring, including services delivered by assistants under general supervision. Remote therapeutic monitoring has traditionally centered on non-physiological data such as adherence, pain, and functional status. Adding contactless circulation data extends what a PT platform can observe without changing the patient's experience, turning a movement-only video session into one that also shows cardiovascular response to that movement.

Cardiac rehabilitation

Cardiac rehab is the clearest fit. Digitally supported home-based cardiac rehabilitation is increasingly found non-inferior to center-based programs for exercise capacity and quality of life, and the 2024 REHAB+ trial is examining mobile cardiac telerehabilitation against center-based care directly. These programs live or die on physiological monitoring during exertion. A platform that can surface heart rate and circulation trends through the camera removes the single biggest barrier to home-based cardiac rehab: the assumption that safe exercise monitoring requires the patient to wear and manage telemetry hardware.

General exercise and post-surgical recovery

Orthopedic recovery, pulmonary rehab, and chronic disease exercise programs all benefit from the same capability. A patient rebuilding capacity after surgery can be pushed appropriately when the clinician sees real circulatory response, and held back the moment readings drift outside a safe band.

Current research and evidence

The evidence base for motion-tolerant contactless monitoring has matured quickly. Beyond the 2.29 bpm yoga result and the 3.41 bpm RMSE from two-stage artifact reduction noted above, 2023 work on recursive motion artifact removal using smartphone camera signals reported a mean absolute error of roughly 3.58 bpm in real-time conditions. Deep learning approaches, including recurrent architectures tuned to suppress motion noise, are consistently outperforming conventional signal-processing pipelines in published reviews of remote heart rate measurement.

On the clinical side, the 2024 wearable-based cardiac telerehabilitation study showing parity with gym-based programs matters even though it used contact sensors, because it establishes that remote exercise monitoring produces equivalent outcomes when physiological data is present. The open question the rPPG literature is now closing is whether that same data can be captured without the sensor. The trend lines from motion-artifact research suggest the answer is increasingly yes for heart rate and circulatory trends, with the usual caveat that contactless methods are decision-support signals rather than diagnostic-grade telemetry.

It is worth stating plainly: camera-based vitals during heavy motion remain an active research frontier, and confidence intervals widen as intensity rises. The responsible product design treats these readings as trend indicators that gate and flag rather than as certified clinical measurements.

The future of telehealth vital signs integration in rehabilitation

The direction is toward sessions where circulation, heart rate, and recovery are visible to the clinician as ambient data, captured passively while the patient simply does their prescribed exercises in front of a camera. Three developments will shape the next phase:

• Motion-robust models trained specifically on exercise video rather than seated baselines, narrowing the accuracy gap between rest and exertion.
• Tighter coupling of vitals streams with billing workflows, as remote therapeutic and physiologic monitoring reimbursement continues to formalize.
• Multi-signal capture, where breathing rate and stress indicators join heart rate to give therapists a fuller picture of exertion and tolerance.

For telemedicine software vendors serving niche rehabilitation markets, the competitive question is shifting from whether vitals can be captured to whether they can be captured during the activity that defines the specialty. Platforms that solve exercise-time monitoring will own a capability that generic video tools cannot match.

Frequently asked questions

Can a webcam really measure my circulation while I am moving? A webcam can estimate heart rate and circulatory trends from subtle skin color changes using rPPG. During movement this is harder than at rest, but 2023 research using motion-artifact-removal techniques reported errors as low as 2.29 to 3.58 beats per minute for moderate activity. Well-built systems suppress unreliable readings during intense motion instead of reporting false values.

How is this different from a fingertip pulse clip or chest strap? Those methods require the patient to own, charge, and correctly wear a device, and contact-based clips in particular tend to fail during movement. Contactless rPPG uses the existing video feed with no hardware, which removes the adherence and onboarding barriers that limit home-based rehab programs.

Is contactless exercise monitoring accurate enough for clinical decisions? Current camera-based vitals are best treated as decision-support and trend data rather than diagnostic-grade telemetry, especially during high-intensity exercise where accuracy margins widen. They are well suited to spotting overexertion, tracking recovery slope, and keeping patients inside a prescribed effort range.

Does video-based vitals capture affect patient privacy? Yes, and it must be handled accordingly. Vital signs derived from video are protected health information subject to the same safeguards as any clinical data, including encryption, access controls, and appropriate consent.

Circadify is building toward this use case directly, with a contactless rPPG SDK designed to add real-time vital signs to any telehealth or telerehabilitation platform using the video feed that is already there, no patient hardware required. Telemedicine software teams serving physical therapy and cardiac rehab can explore the platform demo and SDK documentation at circadify.com/custom-builds.