FAQ
Frequently asked questions
Short answers to the questions laboratories, diagnostics makers, trial teams and data teams ask us, each linked to the detail.
What does NakedSignal do?
NakedSignal measures what a change of analyser, reagent or platform does to results at the clinical decision limits that matter, and hands over evidence that can be re-run. The same method applies to sites and devices in clinical trials and to the inputs of medical AI.
Who is it for?
Clinical laboratories and networks changing or running several analysers; diagnostics makers and EQA or proficiency-testing schemes; and trial sponsors, CROs and medical-AI teams who need site, lab or device effects sized at their thresholds.
What data do you need from a laboratory?
The paired results from the method comparison or parallel run you already do, your decision limits for each analyte, and, if you have it, the repeat-test imprecision of both analysers.
Does our data leave the laboratory?
The program runs on your own computer with no network connection, and only aggregate files leave: the aggregate results, the report and a release statement you sign. Small counts are suppressed. A laboratory can also choose to share coded paired results with us under a written data agreement, or to contribute coded pairs or aggregates to the Analyser-Change Registry. Both are optional, a report never needs either, and your data protection officer decides what is included.
How does the report fit our ISO 15189 verification?
It runs on the paired results you already collect for verification and adds what validation software leaves out: bias at each decision limit, the patients who cross each limit against repeat testing, a re-test band for the transition and a note for clinicians.
Where do the numbers on this site come from?
From runs of our program on open-access published data, each study cited with its licence: 10,579 paired results in 34 method comparisons at 14 study sites, from 12 studies.
Can I check your results myself?
Yes. The change explorer shows every public comparison with its source study and needs no sign-up. A technical validation kit that rebuilds the headline result offline is available on request.
What is MedEval-1?
MedEval-1 is a versioned evaluation standard for medical AI, built by NakedSignal. Its held-out track seals, scores and rotates a test set, and 48 of 48 of its published scores were re-derived by audit from the published manifests.
How much does it cost?
Each offer states its price; ask for a first site report and we will send you the offer with its price. A first look covers up to three analytes and is delivered within ten working days of receiving the data, and its fee is credited in full against a full report.
How do I get in touch?
Use the contact form; we reply by email.