Documenting symptoms with photos and video for the vet
How to film a short, intermittent episode so it is readable in the consult: what to frame, how long to record, what to note, and what never to do.
- Audience
- Pet owners
- Species
- All species
- Scope
- Valid everywhere
The most common problem in a consultation is not the absence of symptoms but their absence at exactly the wrong moment. An animal that coughs, limps, has a strange muscle twitch or laboured breathing at home arrives at the clinic and, in front of the vet, looks perfect. The episode lasts seconds, happens at night, or is triggered by a specific situation that the consulting room cannot recreate. A spoken account is bound to stay vague: the word cough covers very different sounds, and the word fainting spans pictures that call for opposite responses. The camera everyone carries closes much of that gap, but only if the video shows the right things, lasts the right time, and reaches the vet with the few surrounding details that make it readable. Filming at random, by contrast, adds noise and helps no one.
Why a video beats a description
Veterinary medicine often works on events the owner describes and the clinician never sees. A seizure lasting thirty seconds, a collapse on waking, a lameness that appears only after running: these are pictures where the diagnosis shifts on details that words cannot carry. A rhythmic, symmetrical tremor is not the same as a sudden jerk of a single limb, yet both get reported as it was shaking. The recording restores rhythm, symmetry, duration, state of awareness and recovery, all information that steers clinical reasoning before any test.
There is a practical benefit that shows over time too. An episode filmed today becomes a reference six months from now: comparing the new clip with the old one shows whether the picture is worsening, holding steady or responding to treatment. Human memory cannot make that comparison, the file can. That is why it pays to keep videos with the rest of the clinical history rather than scattered in the phone gallery, where they become impossible to find when they are needed.
What to frame for each type of episode
A good video is not a tight close up of the worrying detail. You almost always need the whole animal, filmed from far enough to see posture and all four limbs, plus the context: the floor, the situation the episode starts from, what happens just before and just after. The type of problem decides what the vet will look for in the clip.
| Symptom | Useful framing | Decisive detail |
|---|---|---|
| Cough or abnormal breathing | Chest and flanks from the side, animal still and at rest | Rib cage movement and a clear sound |
| Lameness | Animal walking and trotting on a non slip surface | Which limb bears less weight and in which phase of the stride |
| Seizure or tremor | The whole body, with a voice calling its name | Whether it responds to the call and how it recovers after |
| Collapse or fainting | Recording started as early as possible and kept through recovery | Real duration and state on coming round |
| Repeated odd behaviour | The whole room, with the objects and people present | What triggers it and what makes it stop |
For the resting respiratory rate, one of the most valuable parameters in heart disease, the ideal video is dull: the animal in deep sleep, filmed from the side for at least thirty seconds, with no noise to wake it. What counts is chest movement, not a lively scene. A calm, boring clip here is worth more than a spectacular one.
The technical rules that make the file readable
- Light on the subject: switch on the room or open the window before filming, because in the dark the video is useless and the flash often disturbs the animal.
- Phone horizontal and steady: rest it or hold it with two hands, because a shaky video hides the very rhythm that needs reading.
- Sound on: the noise of a cough, a breath or a whine is clinical data, so do not film on silent and do not talk over the episode.
- Right length: twenty seconds to two minutes is almost always enough, while a very long file transfers badly and dilutes the important moment.
- One episode per file: several short, dated videos beat one long clip where the clinician has to hunt for the right point.
- Medium distance: frame the whole animal, then if needed move closer in a second video on the detail, without losing the overall view.
The surrounding information no video captures
A clip with no data around it stays half a clue. The vet needs to know when it happened, how long it really lasted, how often it occurs, what precedes it and how the animal is for the rest of the day. These notes are written in the moment, not from memory in the waiting room, because as the hours pass the numbers get rounded and the sequences blur.
Note the date and time of the episode
The phone records them in the file metadata, but a written reminder helps reconstruct how often episodes occur over time.
Record the real duration with a timer
Thirty seconds lived through feel like five minutes. Measuring the actual time substantially changes the clinical reading of a seizure.
Describe the context in one line
After running, on waking, during a meal, around loud noises. The trigger steers the diagnosis as much as the symptom itself.
Log the frequency
First time, or the third in a week. An isolated episode and a recurring one lead down different paths.
Keep the videos in one tidy place
A health diary with the date attached to each file avoids the frantic search in the gallery and lets the clinician see the full sequence.
This orderly material also opens the door to remote consultation. Many telemedicine requests resolve better when the vet first receives a couple of clear videos and the surrounding notes, rather than having to reconstruct everything during a video call in which the animal, once again, does not repeat the symptom.
Mistakes that make a video useless
- Heavy digital zoom: it breaks up the image and loses context, so move physically closer if you need a detail.
- Provoking the symptom to film it: triggering a seizure or forcing a painful movement to get the shot is harmful and not ethical.
- Cropping and applying filters: edits change colours and timing and strip the file of diagnostic value.
- Filming only the worst moment: without the before and after, the clinician loses the part that separates the hypotheses.
- Sending dozens of files with no order: quantity does not help, a couple of chosen, dated videos do far more.
- Waiting for the perfect shot during a real emergency: if the animal is unwell now, you call, you do not film.
It is worth remembering, finally, that these tools help to tell the story, not to decide. An app that analyses a clip or suggests hypotheses can guide the urgency of a night, but it remains a support for triage and not a diagnosis. Reading a video, like reading a test, belongs to the vet who knows the animal and its history.
Frequently asked questions
- How long should the video for the vet be?
- In most cases twenty seconds to two minutes is enough. You need long enough to capture the start, the full picture and the recovery, but not so long that the file becomes heavy or the moment that matters gets diluted. For a very brief event such as a muscle twitch, a short clip works, as long as it frames the whole animal. If the event is long, several dated files beat one endless clip that is hard to transfer and to review.
- Photo or video, which is better?
- It depends on the problem. For something static, such as swelling, reddened skin, gum colour or the look of a wound, a sharp, well lit photo is often enough and easier to read. For anything that changes over time, such as cough, lameness, breathing, tremors or behaviour changes, you need video, because it is precisely the rhythm and sequence that carry the information. When in doubt, a few seconds of video still contains the still frames too.
- How do I keep videos so they can be found again?
- The main risk is that the clip vanishes into the phone gallery and becomes impossible to find at the visit. It helps to gather the videos with the rest of the clinical history, with the date and a short note on what they show, so the vet sees the sequence over time and can compare today's episode with one from six months ago. A tidy health diary makes that comparison immediate and saves scrolling through hundreds of images during the consult.
- Can automatic video analysis tell me what is wrong?
- No. A tool that processes a clip or suggests hypotheses can help gauge how much urgency there is, but it stays a triage support, not a diagnosis. Gum colour in a video can be distorted by light, a tremor can have many different causes, and no model knows the animal's full history. The recording exists to bring the vet a faithful piece of information, which they still interpret together with the examination and the tests.
What to do next
When a symptom appears at home and vanishes at the clinic, film it with good light, the phone steady and horizontal, sound on and the whole animal in frame, trying to capture before, during and after the episode. Time the real duration with a timer, note the date, context and frequency, and keep the files in the clinical history with their date, so the vet reads a faithful record and can compare it over time.
Related content
- Read10 min read
Triage is not diagnosis: the real limits of artificial intelligence
A system that takes symptoms and returns an urgency level does one thing: it sorts. It does not look at the animal, palpate it or measure anything nobody measured first. Here is where the line falls.
All species - Read8 min read
Veterinary telemedicine: when it helps and when you need a visit
A screen changes the channel, not the medicine. A remote consult can orient and monitor, but it does not replace hands on the animal when a diagnosis is needed: that line is worth knowing.
All species - Read8 min read
Resting respiratory rate: how to measure it at home
A number counted at home while the animal sleeps beats many impressions. Here is how to measure resting respiratory rate, which value to watch, and when it becomes an alarm.
DogCat