Important safety information
The device output is not a clinical diagnosis. The probability distributions, severity scores and clinical sign quantifications produced by the device are clinical decision support information that must be interpreted by a qualified healthcare professional in the context of the patient's complete clinical picture.
The device must never be used to confirm a clinical diagnosis. Its output is one element of the overall clinical assessment. The healthcare professional retains full diagnostic authority and independent clinical judgement in all cases.
Understanding the device output
The device produces probabilistic outputs, that is, a ranked list of possible conditions with associated probability scores, rather than a definitive diagnostic conclusion. This is a fundamental characteristic of how it operates.
- The device cannot confirm the presence or the absence of a condition. The output is always a probability distribution across ICD classes, never a binary "positive" or "negative" result.
- The probability distribution may be concentrated on a small number of classes or dispersed across many, but it always represents a range of possibilities rather than a single conclusion.
- Severity scores and clinical sign quantifications are measurements intended to support clinical assessment, not to replace it.
- The device analyses what is visible in the image. It has no access to the patient's history, symptoms, physical examination findings or laboratory results, and cannot take into account anything that is not visible in the image supplied to it.
Population and performance variability
The device is intended for use on the patient population declared in Intended use or purpose. Within that population, the performance of the device may vary across demographic subgroups, including Fitzpatrick skin phototypes, age groups and geographic populations that are not proportionally represented in the datasets used to validate the device.
Healthcare professionals should exercise particular clinical judgement in the following situations.
- Assessing patients with higher Fitzpatrick skin phototypes, in whom visible clinical signs such as erythema are less readily discernible in an image.
- Assessing paediatric or geriatric patients, whose presentation patterns may differ from those of the adult population.
- Working in geographic regions or clinical settings whose condition prevalence differs from that of the datasets used to validate the device.
The device output should always be interpreted in the context of the individual patient's clinical presentation. This variability does not alter the safety profile of the device, because the healthcare professional retains full diagnostic authority in all cases, as described below.
The healthcare professional retains diagnostic authority
The device is designed to be used when a healthcare professional chooses to obtain additional information to support a clinical decision. The healthcare professional combines their clinical knowledge, the patient's medical history, the physical examination findings, any laboratory results and the device output to reach a diagnostic and therapeutic decision.
The device must always be used under the supervision of a healthcare professional, who should confirm or validate its output in the light of the complete clinical picture. Where the device output and the healthcare professional's own assessment disagree, the healthcare professional's clinical judgement prevails.
If the device returns an implausible result or fails to return one
If the output does not correspond to what you observe in the patient, or if the device returns an error or no result at all, proceed as follows.
- Do not act on the output. Continue the clinical assessment using your own clinical judgement and the usual standard of care, exactly as you would if the device were unavailable.
- Check the image quality and image domain information that the device returns alongside the clinical information. An image that is out of focus, poorly lit, taken too far from the skin structure or that does not show a skin structure at all can produce an unreliable result. See Domain validator for how the device reports this.
- Retake the image following the guidance in How to take pictures and submit it again.
- If the result is still implausible, or the failure persists, report it to the manufacturer at support@legit.health so that it can be investigated. Describe what you observed and what the device returned.
- If the event meets the criteria for a serious incident, report it as described in UK market access and vigilance for devices placed on the UK market, and to the national competent authority of the country concerned elsewhere.
Contraindications and residual risks
The situations in which the device should not be used, together with the residual risks identified for the device and the recommended course of action for each, are set out in Intended use or purpose. Read that section before using the device.