Grade C/B pharma cleanroom — cap, mask and overgloves verified before access.
The integrity of a grade C/B pharma lot starts in the gowning antechamber: correct cap, mask and overgloves before the airlock. iLEAN Vision verifies every operator in the moment before access and holds the airlock if something does not add up. The quality manager signs off — the room is never contaminated by a badly placed cap.
The gowning procedure is perfect on paper — and the operator enters a hundred times a month.
A grade C/B cleanroom is contaminated by what comes in, not by what goes out. And what comes in are people who have been at the plant for ten years and know the gowning procedure by heart. Even so, at every entry there are three pieces that must be properly in place, with no assumptions:
- The cap covering all the hair, including temples and nape, with no loose strands.
- The mask correctly positioned above the nasal bridge, not below it.
- The sterile overgloves, both on, in the approved model and disinfected before the airlock.
The quality manager knows this, but cannot be at every airlock every time every operator enters — there are hundreds of entries a day across the three shifts. Neither can the delegated supervisor; they look at what they can and sign. The classic system works 99% of the time. That 1% is the deviation that shows up in the GMP audit — and each one drags an investigation, a possible lot re-qualification and a shadow the quality manager does not want over their signature.
iLEAN does not add a fourth procedure — it puts the quality manager's eyes in the antechamber, without them having to be there.
The problem with gowning control in the cleanroom is not a lack of procedure: it is verification that lives in one person who physically cannot be at every airlock. iLEAN acts as the filler that closes that gap, without asking you to change the gowning, the procedure or the quality system. See the category: full IRIS architecture.
Edge sees the cap, the mask and the overgloves before the airlock. Connect reads the approved procedure and the garment catalogue. The agent cross-checks and, if something does not add up, holds the airlock. The quality manager signs off — never the other way round.
The three iLEAN pieces applied to gowning verification in the cleanroom:
- iLEAN Vision (Edge) — a terminal with a camera in the antechamber, fixed at the angle defined by the gowning procedure. The CNN is trained on the approved garment catalogue (cap, mask, sterile overgloves from the validated supplier). It looks at the garment and its correct placement — it does no facial recognition. It works with no network: if the line goes down, Edge keeps verifying against the last synchronised set.
- Connect — captures the current gowning procedure from the document management system (DMS), the approved garments and their lots, and the sterile supplier's catalogue. It also captures what arrives from outside (an overglove supplier change notified by email, an SOP update signed by quality).
- Agent — cross-checks the procedure, the detected garments and their placement. If something does not add up, it shows the operator what to correct (no public display with their name) and alerts the quality manager through Connect. The airlock never starts on its own — the person validates and signs off. And the agent closes the entry's dossier for the auditor.
Manual gowning verification vs. verification with iLEAN Vision
| Aspect | Procedure + supervisor | With iLEAN Vision + Connect + Agent |
|---|---|---|
| Cap check (temples and nape) | Supervisor's glance, assumption | CNN trained on the approved model |
| Mask position | The supervisor's eye | Detection of the mask above the nasal bridge |
| Overglove detection | The operator confirms verbally | Verification of both gloves in the approved model |
| The "supervisor is at another airlock" case | Airlock with no visual check | Constant verification at every entry |
| Dossier for the GMP audit | Reconstructed after the fact, weeks | Dossier per entry, automatic |
| Operator privacy | n/a | No facial recognition; identification by badge |
Impact estimate for your plant — to be validated with your numbers.
The block below is an estimate to be validated with the specific figures of your room. We set it out so the committee has an order of magnitude; we refine it during the diagnostic.
- Pharma plant with 2 grade C/B cleanrooms, two access antechambers, several entries per operator and shift.
- Edge pilot at one antechamber (camera + integration with the procedure's DMS + access control). First expected value within a few weeks: verification genuinely useful for quality.
- Indicative payback between 4 and 9 months, depending on the number of entries and the documented average cost of last year's gowning deviations.
- Expected reduction in gowning deviations detected after the fact of ≥ 30% versus the previous manual control, with a floor that is reasonably defensible before the committee.
- The hard lever is a single avoided GMP deviation: investigation, possible lot re-qualification, product hold, reputational damage before the auditor.
And the quality manager's reasonable doubt
"What if the AI confuses a well-placed mask with a badly placed one?" — hallucination is a problem of free generation, not of anchored tasks. In tasks where the AI simply classifies an image against a catalogue of garments and positions (cap OK, mask above the nasal bridge yes/no), the best models brought the error below 1.5% [1]. And even so, what is critical is never decided alone: the airlock does not start and the quality manager signs off. The three safety rings are there for exactly this.
[1] OpenAI paper "Why Language Models Hallucinate", 2025 — on the reliability of AI in anchored tasks.
What people ask about cleanroom PPE detection in pharma
What does EU GMP Annex 1 require for gowning in a grade C/B cleanroom?
Annex 1 of the EU GMP requires protective gowning to match the grade of the room and the operation. In grade C: clothing that covers the body, dedicated footwear, a hood/cap that covers the hair and, where applicable, a mask. In grade B (and A): a sterile one-piece suit, a hood that completely covers the hair, mask, goggles and two pairs of sterile gloves disinfected before access. The operating rule: the room is contaminated by what comes in, not by what goes out. Every operator enters one and a half times per shift — the classic system (procedure + goodwill) works almost always, and the "almost" is what shows up in the audit.
How does iLEAN Vision verify the cap, mask and overgloves at once?
With an Edge terminal in the antechamber, just before the airlock into the grade C/B area. The camera observes the operator from the angle defined by the gowning procedure, and the neural network (CNN) is trained on the specific set of garments used at the plant — approved cap and mask models, sterile overgloves with the validated supplier's code. It verifies that the hair is completely covered, that the mask is correctly positioned (not below the nose), and that both overgloves are on before the airlock button is pressed.
What happens if an operator arrives with the cap badly placed?
The airlock does not start. iLEAN Vision shows the operator on screen the specific piece that needs correcting (e.g. a visible strand at the right temple, mask below the nasal bridge) without naming the operator on the public display. The system leaves the decision to the quality manager, who validates from their station or from Connect. If the operator corrects it and the verification is OK, they pass through. If not, there is a documented case, with no entries into the room with incorrect gowning. The person signs off.
Does this verification serve as evidence for a GMP / FDA audit?
Yes, that is one of its main returns. Every entry leaves a structured dossier: timestamp, a photo of the operator at the moment of verification, the list of garments detected, the quality manager's human validation and a cryptographic signature of the record. Reconstructing today's entries into a C/B room for an IFS/BRC/FDA auditor is weeks of work; with iLEAN, the dossier is already made. The information governance is designed so the auditor can read it without your quality team losing three working days.
How much does the pilot cost in a pharma plant with 2 access antechambers?
The order of magnitude of an Edge pilot in a pharma antechamber is close to that of any Edge pilot in a regulated plant: an initial investment covering the terminal + camera + integration with the access control and the quality document management system, plus an annual licence. The reasonable payback to present to the committee is a matter of several months — the hard lever is a single avoided GMP deviation (investigation, lot re-qualification, possible product hold), which weighs more than the cost of the system. Send us your room's data and we will send back the estimated ROI within 48h.
Tell us your case and within 48h we will send you the estimated ROI of this AI project for your cleanroom.
We work on your plant's real data, not on ours. Diagnostic with no commitment.
Request estimated ROI in 48h See iLEAN Vision