Smart mirrors for earlier skin cancer detection

A smart mirror for checking early signs of skin cancer in remote clinics of Sri Lanka could turn an informal examination into a structured screening visit. A camera, controlled lighting and decision-support software would help clinic staff document suspicious lesions for review. Its role should be triage and referral, not autonomous diagnosis.

The need is practical. Dermatology expertise and specialist services are concentrated around major cities, while travel from remote areas can delay assessment. Sri Lanka’s strong UV exposure makes consistent sun protection important, and visible warning signs may be noticed late. A connected tool could help trained community clinicians identify which lesions require closer or faster review.

For Australian readers, the proposal offers a relevant digital-health case at home and abroad. Skin cancer awareness is high in Sydney, Melbourne and Brisbane, where people routinely combine outdoor activities with broad-spectrum sunscreen, hats and shade. Connected devices, telehealth and commercial skin checks already form part of the Australian market, while medical-device and privacy requirements shape trusted adoption.

Why early detection needs a local pathway

Remote clinics need a clear entry point rather than a device that simply displays an automated risk score. During a routine consultation, staff can ask about changing or bleeding lesions, examine the relevant area with a calibrated camera and record standardised images. Sinhala- and Tamil-language prompts could support structured history-taking without replacing clinical judgement.

Findings could be grouped into urgent, soon-to-review and routine categories. The clinician would confirm each classification, with teledermatology support available when the image or history is uncertain. Training should cover common presentations in Sri Lankan skin tones, non-healing wounds and changes that require escalation, reducing the risk that an algorithm overlooking subtle lesions.

Designing the clinical workflow

Each visit should begin with informed consent and a clear explanation of what is photographed. The mirror can guide the patient toward the affected area, while a fixed camera and standardised lighting produce comparable images. Automatic quality checks can identify blur, poor framing or incorrect magnification before an image reaches the clinician or specialist.

At the end of the encounter, staff can attach images to the patient record and send selected cases for remote review. Results should support an explicit plan: immediate referral for concerning lesions, an appointment within a defined period, short-interval monitoring or reassurance with safety advice. The system should never issue a diagnosis without review by an appropriately trained health professional.

Engineering for clinics and communities

The device must function within the realities of a remote practice, including intermittent connectivity and constrained power budgets. It should capture images offline, synchronise them when a secure connection becomes available and retain a clear audit trail. A low-power mode, optional solar charging and a replaceable dermatoscope attachment would improve reliability while keeping operating costs manageable.

Physical design also affects trust. A privacy shutter, visible recording indicators and limited automatic image retention can reassure patients. The interface should be usable by older clinicians and patients with limited digital literacy, while infection-control procedures must cover camera surfaces, mirrors and removable attachments. Regular calibration, maintenance and replacement plans are as important as the initial purchase.

Evidence, privacy and regulatory safeguards

A pilot should compare smart-mirror assessment with the clinic’s existing care pathway rather than treating the technology as proof of benefit. Prospective evaluation should measure clinically meaningful sensitivity, false-negative findings, referral agreement, completed referrals, consultation time, system uptime and patient understanding. Images should represent different regions, skin tones and lesion types, and dermatologists should independently review a representative sample.

Health images are sensitive information. Consent should distinguish clinical care from research, commercial use and model training. Access should be role-based, data encrypted, retention periods defined and audit logs maintained. A Sri Lankan deployment would need to align image handling with the Data Protection Act No. 1 of 2022, applicable health-research requirements and the policies of participating institutions.

Australian involvement brings additional regulatory considerations. Organisations operating under the Privacy Act 1988 must consider the Australian Privacy Principles, while connections with My Health Records require attention to the My Health Records Act 2012. If the software entered the Australian market, sponsors would need to determine its status under the Therapeutic Goods Administration’s medical-device framework and meet the applicable evidence, quality and post-market requirements.

Australian expertise and market fit

Australian partners can contribute expertise in dermatoscopy, clinical workflow design, telehealth, low-bandwidth systems and evaluation of digital health services. Australia’s experience with remote healthcare can be useful, particularly for clinics operating beyond major metropolitan centres. Commercial skin-check services and consumer wearable devices also provide a reference for how patients expect accessible screening technology to work.

That expertise should support locally led implementation rather than replace Sri Lankan clinical priorities. Sri Lankan clinicians should determine referral protocols, language, acceptable data practices and procurement priorities. Australian organisations can strengthen training, independent evaluation and financing, while local clinics retain control over patient care and the technology purchased for their communities.

A staged path from pilot to scale

The first stage should establish a baseline across several representative clinics before installing hardware. A limited pilot can then test image quality, referral completion, staff workload and patient acceptance. Expansion should depend on predefined safety and performance measures, an affordable maintenance model and a referral system capable of absorbing additional cases.

The investment case should include the full cost of ownership: devices, connectivity, maintenance, training, software and specialist review time. Blended support from development agencies, public health programmes, telecommunications partners and philanthropic funding could make the service sustainable without charging patients per image. Relevant stakeholders can compare regional digital-development priorities through the ICTD-ASP launch events before joining a public concept note for the pilot.

The next concrete step is to convene Sri Lankan clinicians, regulators, rural clinic representatives, an Australian technical adviser and a local data-protection officer, then select representative sites and approve shared consent, referral and evaluation protocols before procurement.