How to track hair progress with photos: a clinician's method

Hair changes slowly. A patient who is genuinely regrowing and a patient who is quietly getting worse can look identical in the mirror for months — which is why memory and a single phone snapshot are poor tools for a decision that plays out over half a year. A standardized photo series turns that slow, invisible drift into something you can measure and show. The catch is in one word: standardized. A pile of photos taken at different distances, under different light, of slightly different areas is not a record of hair change — it is a record of your photography. This guide covers the method clinicians use to make hair photos comparable, the intervals and views that matter, when a global photo is enough and when you need to count hairs, and how we at Imaginos structure the whole thing so the series stays honest.

Why photograph hair progress at all?

Because the alternative — subjective recall — fails exactly when it matters. Hair-loss treatments such as minoxidil and finasteride typically need at least six months before a visible effect emerges, and the change is gradual enough that neither patient nor clinician reliably perceives it from visit to visit. That perception gap has real consequences: in one survey, before viewing comparison photos, 34% of patients believed their hair loss was "worse" or "much worse" and 81% reported alopecia-related anxiety; after reviewing standardized photos with their physician, 53% recognized improvement, anxiety dropped, and the correlation between how patients and clinicians graded severity roughly doubled.^[1]

That is not a cosmetic benefit. Standardized comparison photography has been associated with better follow-up adherence — patients who underwent regular photographic assessment were significantly less likely to be lost to follow-up for both androgenetic alopecia and alopecia areata, with the effect strongest in the first year of treatment.^[2] Given that adherence is the single biggest determinant of whether a slow-acting hair therapy works, "the patient can see it is working, so they keep going" is a clinical outcome, not a nicety.

What actually makes two hair photos comparable?

The governing principle in clinical photography is blunt: fix everything that is not the hair. Standardization matters most for before/after and follow-up images, and it spans camera parameters, patient position, lighting, and background.^[3] Concretely, that means holding constant:

• Device and settings. Same camera across the series; lock the zoom and turn off filters, "portrait"/beauty modes, and heavy auto-brightness, which silently alter the very thing you are trying to measure.^[4] A modern smartphone is fully capable here — the priority is consistent technique, not expensive gear.^[3]
• Distance and angle. Reproducible camera distance with the lens held perpendicular (90°) to the scalp; mark a fixed position or use a reference so you are not eyeballing it each visit.^[5] Getting physically closer distorts the image — use digital zoom instead of moving the lens in.^[4]
• Lighting. Even, consistent illumination without hot spots or shadows; consensus guidance recommends an even light source, angled or oblique to the surface, and a clip-on ring light at ~45° is a cheap way to get uniform light and avoid glare, especially on darker skin and scalp.^{[5, 4]}
• The region and the order. Same part line, same regions, captured in the same sequence every visit.

The point of all this discipline is a single question you should be able to answer honestly at each visit: if the hair were completely unchanged, would these two photos look the same? If styling, wetness, or lighting could account for the difference you are seeing, the setup was not standardized enough.

Which views should you capture?

Hair-loss trials converged long ago on a small set of standardized global-photography views, a discipline whose foundations were laid by Canfield in the 1990s^[6] and adopted across finasteride and other pivotal studies.^[7] Guideline bodies describe the standardized set as the vertex, mid-pattern, frontal, and temporal regions, captured with constant view, magnification, and lighting for follow-up.^[8] A pragmatic clinical protocol reduces this to what a busy practice can actually repeat, with presentation-specific variants for the main pictures (pattern loss, frontal fibrosing/traction alopecia, and patchy loss such as alopecia areata).^[9]

For most androgenetic cases, four views cover it:

1. Frontal / hairline — recession and frontal density.
2. Mid-scalp — the central part; the region most scales key on.
3. Vertex / crown — often where change shows first and clearest.
4. Temporal — completes the picture and anchors alignment.

A practical, low-cost fix worth knowing: a brief staff-training intervention on when to photograph and which views to capture roughly doubled the rate of correct, standardized photos in one academic clinic (from 32% to 66%).^[10] The technique is learnable; the failure mode is almost always workflow, not equipment.

How often should you take them?

Match the interval to how fast hair actually changes. Because response takes six months or more, most treatment courses are best served by a baseline plus follow-ups every 3 to 6 months. Tighter cadences mostly capture noise; wider ones risk missing the window where a non-responder should be switched. The rule that matters more than the exact number: pick an interval and keep it — even spacing is what makes the trend readable rather than a set of one-offs.

When is a global photo enough, and when do you need to count hairs?

A standardized global photo answers coverage: is the area of visible thinning growing or shrinking? For decades that has been scored on a seven-point scale (from −3 "greatly decreased" to +3 "greatly increased"), a method with documented reproducibility — trained raters reach correlations around 0.76–0.8 on paired before/after images.^[11] If your question is "is overall coverage better or worse," well-standardized global photos, ideally read against a validated severity scale, are the workhorse.

Those scales are worth knowing because they turn a picture into a defensible grade:

• Norwood–Hamilton for male pattern loss (stages I–VII).^[12]
• Ludwig (three grades) and the Sinclair five-point scale for female pattern loss.^[12]
• SALT (Severity of Alopecia Tool) — the percentage of scalp with terminal-hair loss, the standardized measure for alopecia areata.^[13]

When the question shifts to what the hair itself is doing — density, caliber, the shift from terminal to vellus hair — a global photo cannot resolve it, and you move to trichoscopy (dermoscopy of the scalp, a term coined by Rudnicka and Olszewska in 2006) or a phototrichogram. Trichoscopy lets you track the hallmark signals of androgenetic alopecia over time: hair diameter variability (present in ~94% of AGA patients in a systematic review), the proportion of vellus hairs (~66%), and the frontal-versus-occipital difference in density and single-hair follicular units.^[14] It is explicitly useful for evaluating treatment response photographically at each follow-up.^[15]

For hard numbers — hair density in hairs/cm², anagen/telogen ratio, terminal/vellus ratio — the phototrichogram is the long-standing non-invasive reference method, and automated tools like TrichoScan compute these from a standardized close-up.^{[16, 17]} Here honesty demands a caveat we will not hide: automated counting is convenient but not infallible. Independent evaluation found TrichoScan can overestimate total hair density and produce vellus-hair percentages that do not track clinical severity, with repeated detection errors where hairs cross, overlap, or vary in thickness — the authors concluded the "validated and precise" framing was an overstatement and that manual correction is often needed.^[18] The practical takeaway: a number is only as trustworthy as the standardized image under it, and quantitative output should support the clinician's read, not replace it.

How do you keep the series honest over months and years?

This is where most photo archives quietly fall apart: six months later, nobody can find the right earlier photo of the exact same spot, so the comparison silently drifts. Reproducibility of the target area is so important that quantitative protocols traditionally used a tattoo to mark the measured spot so the same area is measured every time.^[8] Most clinics will not tattoo patients — which is exactly the problem software should solve.

Here is how we structure it at Imaginos, and why:

• Regions as the organizing unit. Instead of a folder of loose images, documentation is organized by anatomical region. "The crown across the last year" is a first-class thing you open, not a search you perform.
• Two families of series — and both are tracked. A macro series (standardized global views, no marker) and a marker series (a fixed, revisitable target area for trichoscopy/phototrichogram) coexist. Crucially, we follow the macro photos longitudinally too — the wide standardized views, not just the dermatoscopic close-ups. Systems built around a single dermatoscope often track only the close-up, but coverage change lives in the macro view.
• A marker replaces the tattoo. A flexible marker pins the target area digitally, and the same marker can appear across multiple photos, so "the same spot" is enforced by the software rather than by ink or luck.
• Protocols are yours, not fixed. Rather than locking you into one vendor's protocol, capture protocols are customizable — you encode your views, regions, and cadence — and free (unprotocolled) captures are still supported for the messy real world.
• Multi-device capture. A single capture session can be shared across the devices you already own — a phone for the global views, a dermatoscope-equipped device for the trichoscopy close-up — synced into the same patient record in real time.

A minimal protocol you can start today

1. Baseline before treatment. Frontal, mid-scalp, vertex, temporal — standardized as above. No baseline, no valid comparison later.
2. Lock the setup. Same device, locked zoom, no filters; perpendicular to the scalp; consistent, even light; a marked distance.^{[4, 5]}
3. Same regions, same order, every visit.
4. Interval of 3–6 months; give it at least 6 months before judging response.
5. Add a trichoscopy target area (marked and revisitable) when you want density and caliber, not just coverage.
6. Review side by side with the patient — that conversation is where adherence is won.^{[1, 2]}

The honest limits

Photos and their derived scores document and track visible change; they are not a diagnosis. Clinical interpretation belongs to a qualified professional, and automated counts should be read critically, not trusted blindly.^[18] Imaginos is clinical documentation and analysis software — not a medical device — and does not make diagnoses. What it does is make the thing that was always slow and invisible finally measurable, in the language your clinic works in, without buying a camera the size of your treatment room.