Request a demo
Request a demo

Home  /  Technology

Three streams, one record

Biometrics, behavior and environment — measured continuously and reconciled on one timeline. Each stream runs on hardware or methods validated in peer-reviewed work, and every claim on this page is cited.

Stream one · the collar

The collar, validated

The biometric layer is built on PetPace™ collars — the smart-collar most validated in the peer-reviewed veterinary literature.1234 The collar continuously records pulse, respiratory rate, heart rate variability, temperature, activity, and body position.

Measured against the clinical standard

In a thesis study at the University of Veterinary Medicine Vienna, the PetPace™ collar was compared against the Holter-ECG, the clinical reference standard for cardiac monitoring.1 The heart rate measured by the PetPace™ collar tracked the Holter closely. Why it matters: the PetPace™ collar’s sensing layer has been measured against the instrument a cardiologist would use. The limits of that comparison are the reason Panoptic reads every trend against the animal’s own history rather than treating any single value as diagnostic.

Correlated with a validated activity monitor

In the first evaluation of the PetPace™ collar, its activity readings correlated with the Actical monitor, an accelerometer already validated as a measure of movement in dogs.3 Why it matters: the activity signal tracks a known, independent reference rather than standing alone.

It discriminated a disease state

In a 44-dog study, the collar’s output distinguished healthy dogs from dogs with osteoarthritis pain: OA-pain dogs showed lower activity and reduced heart-rate-variability indices.4 Why it matters: the collar detected a difference tied to a clinical condition, not merely a number.

Each animal is its own baseline

A resting heart rate that is ordinary for one dog is a warning in another. Panoptic reads every PetPace™ collar reading against that animal’s established pattern, so a meaningful change surfaces where a population average would hide it.

Stream two · the cameras

The cameras, integrated

Computer vision validates collar readings and provides visual context for posture, gait, and behavior. Panoptic’s video layer is built on deep-learning pose estimation and automated gait analysis — both established, peer-reviewed methods in veterinary science.

Gait analysis detects lameness

Machine-learning gait analysis identified every clinical lameness case, and some subclinical cases, in a heterogeneous population of dogs.5 Why it matters: a limp is often the first visible sign of pain or injury, and it can be read from movement alone.

It works on many animals at once

A deep-learning pose-estimation system scored posture and gait for multiple cattle simultaneously within a single camera’s field of view, at 94–100% accuracy.6 Why it matters: this is the technical basis for holding each animal in view, and handing it off camera to camera, in a room full of animals.

Behavior and pain indicators from video

Video-based deep-learning models estimate pain indicators in dogs from pose and movement over short clips, using datasets annotated by veterinary professionals.7 The field is surveyed in Frontiers in Veterinary Science.8 Why it matters: video becomes a record of how an animal moved and behaved, not just that it was present.

Findings attach to the right animal

The cameras use Panoptic’s proprietary software to pair the videos to the PetPace™ collars, so the animal in frame is a named individual rather than an unidentified animal. A behavioral finding attaches to the same record as that animal’s vital signs.

Stream three · the environment

The environment, measured

Panoptic measures twelve conditions of your facility — temperature, sound, air exchange, humidity, ammonia, carbon monoxide, light cycle, radon, volatile organic compounds, particulates, outdoor air quality, and formaldehyde — continuously, at animal level, against thresholds drawn from published veterinary standards and federal exposure limits. No one else monitors all twelve. Twelve conditions measured separately are twelve numbers; interpreted together, and against the animal’s own physiology, they produce what no single reading can.

1 · Temperature

The only condition that can harm a healthy animal within a single afternoon. Dogs that overheat while confined have six times the odds of a fatal outcome compared with dogs that overheat through exercise.9 Standard: 64–80°F (ASV)10; federal bound 45–85°F, not exceeded beyond four consecutive hours.11 How: at animal level, with time past threshold tracked as a running clock.

2 · Sound

The most extreme condition in any kennel, and the least measured. Peaks regularly exceed 100 dB.12 One shelter instrument saturated at 118.9 dBA and never recorded the true peak.13 All fourteen kenneled dogs in one study showed measurable hearing change at six months.14 Standard: 85 dBA federal action level.15 How: continuous A-weighted measurement with running average and peak capture.

3 · Air exchange

The principal modifiable control on airborne disease transmission, and respiratory illness is the dominant health problem in group-housed animals. A failed fan or blocked diffuser produces no visible signal. Standard: 10–20 air changes per hour (ASV)10; federal workplace cap 5,000 ppm CO₂.16 How: continuous CO₂ measurement as a real-time proxy for ventilation performance.

4 · Relative humidity

Humidity governs whether an animal can shed heat, and how long pathogens persist in the air. Airborne bacteria and virus survival is minimized between 40 and 70%.17 Standard: 30–70% acceptable, 40–60% optimum.10 How: paired with temperature at the same point, so the system evaluates thermal load rather than temperature alone.

5 · Ammonia

Accumulates from waste and is heavier than air, so it pools low — in the space the animal occupies, not the aisle a person walks. Continuous exposure at 20 ppm produced respiratory-tract damage and significantly increased infection rates.18 Standard: NIOSH 25 ppm over eight hours, 35 ppm short-term.19 How: at animal level, inside the housing space — placement is the entire point.

6 · Carbon monoxide

Gas dryers, water heaters, furnaces, and attached vehicle bays are standard equipment in this industry. Carbon monoxide is colorless and odorless; no facility detects it without an instrument. Standard: NIOSH 35 ppm time-weighted, 200 ppm ceiling.19 How: continuous electrochemical detection in every monitored space, with escalation tiers rather than one alarm point.

7 · Light cycle

Light governs sleep, and sleep governs recovery. A light left on overnight is among the most common and most correctable welfare lapses in this industry, and nothing currently catches it. Standard: lighting consistent with natural duration and intensity.10 How: against the facility’s own schedule, alerting on light during designated dark hours rather than a brightness number.

8 · Radon

Radon enters through the slab and settles in the lowest occupied space, which in a boarding facility is the floor the dogs sleep on. Across ten veterinary colleges, dogs and cats in the highest-radon counties showed roughly twice the rate of primary lung tumors as animals in lower-radon counties.20 Standard: EPA action level 4 pCi/L.20 How: continuous measurement at animal level, where the gas collects.

9 · Volatile organic compounds

VOCs come off the things a facility uses daily — disinfectants, floor cleaners, deodorizers, new flooring and bedding. In pets already carrying bronchial or lung disease, higher indoor VOC levels track with measurable change in the lower airway.21 Standard: no federal limit exists for indoor total VOCs; the controls are source management and ventilation. How: continuous total-VOC measurement at animal level, alerting on a rise above the facility’s own established baseline.

10 · Particulates

Dander, dried waste, bedding fiber and foot traffic keep fine particles suspended in any room full of moving animals. In a study of 348 dogs and cats, cats with respiratory disease lived in homes with significantly higher PM2.5 than cats without it, and cats in homes above 35 µg/m³ were more likely to have respiratory disease.22 Standard: 35 µg/m³ PM2.5.22 How: continuous PM2.5 measurement at animal level, not at the ceiling where the air is cleanest.

11 · Air quality index

The only condition on this list that arrives from outside. Wildfire smoke, ozone and regional pollution decide whether the yard is usable at all, and the AVMA is explicit that animals with cardiovascular or respiratory disease are at particular risk whenever air quality is poor.23 Standard: EPA Air Quality Index — above 100 is unhealthy for sensitive groups, above 150 unhealthy for everyone.24 How: the outdoor index for your location, pulled continuously and read against indoor readings, so the system tells you which days the dogs should stay in.

12 · Formaldehyde

Released by pressed-wood fixtures, new flooring, adhesives and some disinfectants, and it keeps off-gassing for months after a build-out. It is a recognized carcinogen, which is why the health-based limit sits nearly fifty times below the enforceable workplace one. Standard: OSHA 0.75 ppm over eight hours, 2 ppm short-term, action level 0.5 ppm; NIOSH recommends 0.016 ppm.25 How: continuous measurement at animal level, with the alert set at the NIOSH figure rather than the OSHA ceiling.

Why three streams

Full context & situational awareness

Three streams supply each other’s context. A rising heart rate in a room that just gained six degrees is heat, not illness. The same rise in a quiet, temperate room, in a dog that has stopped moving, is something to look at now. Panoptic correlates all three across the full record, so a change in an animal is always read against the conditions it was living in at the time.

Sources

Every claim is cited

Biometric and video claims trace to peer-reviewed veterinary and computer-vision literature; environmental thresholds trace to published standards of care and federal exposure limits.

  1. Görisch E. Comparison of the 'PetPace' smart health-monitoring collar to a Holter-ECG in terms of accuracy of heart rate and heart rate variability measurement. Thesis, University of Veterinary Medicine Vienna, 2020. Read the thesis
  2. Scheinowitz M, Lascelles BDX. Comparison of pulse rate measured by the PetPace collar against electrocardiography and pulse oximetry in dogs and cats. Tel Aviv University Department of Biomedical Engineering with North Carolina State University College of Veterinary Medicine, 2017. Reported by the manufacturer; not published in a peer-reviewed journal. Announcement
  3. Belda B, Enomoto M, Case BC, Lascelles BDX. Initial evaluation of PetPace activity monitor. The Veterinary Journal 2018;237:63–68. doi:10.1016/j.tvjl.2018.05.011
  4. Rowlison de Ortiz A, Belda B, Hash J, Enomoto M, Robertson J, Lascelles BDX. Initial exploration of the discriminatory ability of the PetPace collar to detect differences in activity and physiological variables between healthy and osteoarthritic dogs. Frontiers in Pain Research 2022;3:949877. doi:10.3389/fpain.2022.949877
  5. Mitchell L, Scott M, Häusler K, Adami C, Sutcliffe M, Allen M. Machine learning-assisted gait analysis for lameness detection in heterogeneous dog populations. The Veterinary Journal 2026;318:106743. doi:10.1016/j.tvjl.2026.106743
  6. Russello H, van der Tol R, Kootstra G, et al. Deep learning pose estimation for multi-cattle lameness detection. Scientific Reports 2023;13:4425. doi:10.1038/s41598-023-31297-1
  7. Zhu H, Salgırlı Y, Can P, Atılgan D, Salah AA. Video-based estimation of pain indicators in dogs. 2023 11th International Conference on Affective Computing and Intelligent Interaction (ACII). IEEE, 2023. arXiv:2209.13296
  8. Phelan B, Mc Nally T, Cuddy L, Lacey G. Automatic gait analysis in canines using computer vision. Frontiers in Veterinary Science 2026;13. doi:10.3389/fvets.2026.1729697
  9. Hall EJ, Carter AJ, O’Neill DG. Dogs Don’t Die Just in Hot Cars — Exertional Heat-Related Illness (Heatstroke) Is a Greater Threat to UK Dogs. Animals 2020;10(8):1324 (odds ratio 6.06, 95% CI 2.13–17.22). doi:10.3390/ani10081324
  10. Association of Shelter Veterinarians, Guidelines for Standards of Care in Animal Shelters, 2d ed., 2022, §§ 4.6–4.8. Facilities chapter (PDF)
  11. 9 C.F.R. § 3.2 and subsections (a)–(c). eCFR
  12. Sales G, Hubrecht R, Peyvandi A, Milligan S, Shield B. Noise in dog kennelling: is barking a welfare problem for dogs? Applied Animal Behaviour Science 1997;52:321–329. doi:10.1016/S0168-1591(96)01132-X
  13. Coppola CL, Enns RM, Grandin T. Noise in the animal shelter environment: building design and the effects of daily noise exposure. Journal of Applied Animal Welfare Science 2006;9(1):1–7. PubMed 16649947
  14. Scheifele P, Martin D, Clark JG, Kemper D, Wells J. Effect of kennel noise on hearing in dogs. American Journal of Veterinary Research 2012;73(4):482–489. doi:10.2460/ajvr.73.4.482
  15. 29 C.F.R. § 1910.95(c), occupational noise exposure. eCFR
  16. 29 C.F.R. § 1910.1000, Table Z-1, OSHA permissible exposure limits (carbon dioxide 5,000 ppm 8-hour TWA).
  17. Arundel AV, Sterling EM, Biggin JH, Sterling TD. Indirect health effects of relative humidity in indoor environments. Environmental Health Perspectives 1986;65:351–361. doi:10.1289/ehp.8665351
  18. Anderson DP, Beard CW, Hanson RP. The Adverse Effects of Ammonia on Chickens Including Resistance to Infection with Newcastle Disease Virus. Avian Diseases 1964;8(3):369–379. doi:10.2307/1587967
  19. NIOSH Pocket Guide to Chemical Hazards, recommended exposure limits: ammonia 25 ppm TWA, 35 ppm STEL; carbon monoxide 35 ppm TWA, 200 ppm ceiling.
  20. Fowler BL, et al. Ecological level analysis of primary lung tumors in dogs and cats and environmental radon activity. Journal of Veterinary Internal Medicine 2020;34(6):2660–2670. doi:10.1111/jvim.15936
  21. Lin CH, Lo PY, Wu HD. An observational study of the role of indoor air pollution in pets with naturally acquired bronchial/lung disease. Veterinary Medicine and Science 2020. doi:10.1002/vms3.231
  22. Lin CH, Lo PY, Wu HD, Chang C, Wang LC. Association between indoor air pollution and respiratory disease in companion dogs and cats. Journal of Veterinary Internal Medicine 2018;32(3):1259–1267. doi:10.1111/jvim.15143
  23. American Veterinary Medical Association. Wildfire smoke and animals. AVMA
  24. U.S. EPA and partners. Wildfire Smoke: A Guide for Public Health Officials, revised 2021 (Air Quality Index categories). AirNow (PDF)
  25. 29 C.F.R. § 1910.1048, Formaldehyde (PEL 0.75 ppm TWA, 2 ppm STEL, action level 0.5 ppm); NIOSH REL 0.016 ppm TWA. eCFR
A Neapolitan Mastiff wearing the PetPace™ collar against the Panoptic field

Learn how you can benefit today

Request a demo

Whether you run a daycare, a boarding facility, or both, we will show you Panoptic working in a facility like yours.

Request a demo