Postural Analysis: Why Capture Standards Decide Whether Two Records Can Be Compared

Postural analysis is only as useful as it is repeatable, and repeatability is decided by the habits of the person holding the camera rather than by the camera. Two records taken a year apart can be compared only if the conditions were the same, and conditions are usually recorded by accident if at all.

We build systems that combine camera-based posture capture with pressure measurement, and the most common quality problem we see is not hardware. It is a series of small variations in stance, marking and camera position that make a follow-up comparison meaningless.

This article sets out the capture standards we would recommend, why each one exists, and how to store the conditions so that a record remains usable long after the session.

What postural capture is qualified to describe

A posture record describes the position of the body in a defined standing stance: the alignment of the leg, the position of the knee relative to the foot, the level of the pelvis and the posture of the trunk. It describes a moment, in a stance, from a viewpoint.

It does not describe movement, and it does not describe cause. A record that shows the knee tracking medially is a record of a position, not an explanation of why the position occurs, and the difference matters when the record is shown to a client.

It also does not describe load. That is the pressure layer’s job, and the two records together are more useful than either alone because they answer different questions about the same moment.

Stating these limits at the start makes the rest of the standards easier to justify, because every standard exists to protect the one thing the record can honestly claim: that it describes a defined position, repeatably.

It is worth agreeing, in writing, what a posture record will be used for in your practice: explanation to the client, comparison across visits, or documentation. Each use tolerates a different level of variation, and the standard should be set by the strictest one you intend.

Stance, marking and the standing position

The stance has to be defined and reproduced. Feet at a comfortable width, weight even, gaze forward, arms relaxed. The exact wording matters less than the consistency, which is why we recommend writing the instruction down and saying it the same way every time.

Marking the floor is the simplest way to make the stance reproducible. Two marks for the feet, one for the gaze target, and a mark for the camera position. Marks wear out; a photograph of the marks in place, stored with the room documentation, makes them recoverable.

Footwear state belongs in the same standard. Barefoot and shod captures describe different positions, and a series that mixes them is a series that cannot be compared. Record the state rather than assuming it.

For clients who cannot stand unsupported, define a supported variation and record it as a variation. A record that says supported is usable; a record that hides the support is a trap for whoever reads it next.

The stance instruction should be short enough to repeat without reading. Four elements are enough: feet at a comfortable width, weight even, look at the mark, arms relaxed. Anything longer is forgotten by the second client of the day, and anything shorter leaves room for interpretation. It is worth testing the instruction on a colleague and asking them to repeat it back, because the version that feels obvious to the person who wrote it is often ambiguous to everyone else. The instruction is part of the instrument, and it deserves the same care as the camera position.

Camera position and the geometry of the view

Camera height, distance and angle decide what the record shows. A camera at hip height sees a different alignment from one at chest height, and a camera at an angle introduces projection errors that are easy to mistake for posture.

The practical standard is to place the camera where it will stay, record that position, and never move it for a session. If the room requires the camera to be moved, the position should be reproducible: marks on the floor, a bracket with a stop, or a dedicated mount that is not used for anything else.

Lighting is part of the geometry. Strong side light produces shadows that the software may read as edges, and backlight produces silhouettes that lose detail. A room with a window on one side and a lamp on the other will produce records that differ by time of day.

Ask whether the system includes a calibration step for the camera position. A system that assumes a stable geometry and checks it is easier to keep honest than one that allows arbitrary placement without recording it.

A simple template taped to the wall behind the camera, showing the stance and the gaze point, is more effective than a paragraph in a manual. Staff follow what they can see, and a room that shows the standard keeps it.

It is worth measuring once: the distance from camera to the standing mark, and the height of the camera. Two numbers, written on the wall, allow the position to be restored after a room is painted or a shelf is moved. Without them, restoration is guesswork, and a record taken after guesswork is a record with an unknown condition.

What to store with every record

Six fields make a posture record interpretable: date and time, stance instruction, footwear state, camera position reference, operator and any variation used. Six fields, filled in as the session happens.

The operator field is not bureaucratic. Different operators produce subtly different records, and knowing who took a record helps explain a difference that would otherwise be attributed to the client.

The variation field is the one that saves arguments. A record taken with a supported stance, or after a period of sitting, describes a different condition from a standard standing record, and the note is what allows a comparison to be qualified rather than abandoned.

Store the record itself as well as the report. A report is a summary written for a purpose; the underlying record allows a different question to be asked later. Systems that export the record preserve that option, and it is worth asking about before purchase.

Storing the conditions is not the same as storing the data. A record can be complete and still be uninterpretable, because the conditions that made it are absent. The six fields we recommend take under a minute to complete if they are part of the session rather than an afterthought, and the practical way to make them part of the session is to put them on the same screen as the capture. Systems that separate the note-taking from the recording produce records without notes; systems that keep them together produce records that can be read a year later by someone who was not in the room.

Reading posture alongside pressure

The two records are most useful when read together and described separately. Posture describes alignment in a stance; pressure describes load in that stance or during movement. Neither explains the other, and the conversation is strongest when both are on the screen and the professional moves between them deliberately.

A useful sequence is to start with posture, because it is the record clients recognise most easily, then move to pressure to show how load sits relative to what they have just seen. That order produces understanding; the reverse order produces a map that the client cannot connect to their own body.

Avoid combining the two into a single score. Composite indices are attractive because they look decisive, and they obscure exactly the information a professional needs: which record showed what, and in which condition.

The reading should end with a sentence the client can repeat. If the two records cannot be summarised in one sentence, the reading is not finished, and adding more panels will not finish it.

It also helps to keep a reference record: one staff member, captured under the standard conditions, stored as the baseline. When records start to look different for no clinical reason, the reference is what shows whether the room has drifted.

Reading posture and pressure together works best when the professional narrates the sequence out loud. Saying first what the posture record shows, then what the pressure record adds, keeps the two separate in the client’s mind and prevents the impression that one image explains everything. That narration is a skill, and like any skill it improves with a written structure to follow.

Where the standards break and how to notice

Standards break quietly. A camera nudged to reach a cupboard, a stance instruction shortened on a busy day, a client measured in socks because the floor was cold. Each is reasonable in the moment and each makes a later comparison unreliable.

The way to notice is a periodic check: record the same staff member twice, a week apart, with the standard routine, and compare. If the records differ more than expected, something in the routine has drifted, and the drift is usually in the room rather than in the software.

A second check is the marks. Photograph them quarterly. Marks that have been moved, covered or worn away are the most common cause of a slow change in records that nobody can explain.

These checks take minutes and they protect the only claim a posture record can make. A practice that runs them can trust a year-over-year comparison; a practice that does not is comparing two different rooms.

Drift is usually gradual, which is why it is invisible in a single session and obvious in a year. A quarterly check is enough to catch it, and a written note of the check is enough to prove that the standard is being maintained.

It is worth deciding what a difference between two records would mean before you have one. If the standard is working, a difference should be attributable to the client, to a change in the shoe, or to a deliberate change in the routine that was documented. If none of those explanations fits, the standard has drifted and the records should not be compared until it is restored. Writing that decision rule down turns an ambiguous situation into a manageable one, and it gives a practice a reason to check its marks and its camera position rather than discovering the problem through a comparison that does not make sense.

Training and the second operator

A capture standard that lives in one person’s habits is a standard that leaves when that person does. Write it down, train a second operator and test that the second operator produces a comparable record.

The test is simple: both operators record the same client on the same day, and the records are compared. Differences that exceed the expected variation indicate a step in the routine that has not been transferred, and that step is usually the stance instruction.

Repeat training matters for the same reason. We provide remote repeat training for new staff, because a clinic that hires twice a year would otherwise retrain twice a year at the cost of a visit.

The outcome of good training is unglamorous: records that look alike, comparisons that hold, and a professional who can say what changed between two visits without hedging. That is what the instrument was bought for.

The standard is worth reviewing once a year, in the same way a practice reviews its appointment template. Rooms change, staff change and equipment moves, and a standard written for last year’s layout quietly becomes a description of a room that no longer exists. An annual review, with the camera measured and the marks photographed, keeps the record honest for another year.

Frequently asked questions

What we would do next

If you are setting up a capture standard, the X5 foot, ankle and posture system page describes what a session records, the professional services page covers training, and the FAQ page states warranty, support and transit in writing. Tell us your room layout and we will describe the camera position we would recommend.

For related reading, see what the sensor layer adds to a standing assessment and one session, two records.

This article is intended for professional and educational reference. It is not medical advice. Foot assessment equipment produces measurements and images; clinical interpretation and any decision about care remain the responsibility of a qualified professional. Certification scope is limited to the documentation we hold: European Union certification and ISO quality system certification.