I used AI to spot my skin cancer – it saved my life (www.independent.co.uk)

When Lynsey Robertson, an office manager helping to roll out a new AI skin‑cancer screening tool at her GP surgery, scanned a long‑ignored mole out of curiosity, the system flagged it for urgent review. That same day she was referred under the NHS two‑week suspected cancer pathway; surgery and follow‑up confirmed malignant melanoma and a successful wider excision. Robertson credits the AI pathway with catching a lesion she might have dismissed, illustrating how automated triage can accelerate diagnosis and potentially save lives. The tool, developed by Skin Analytics, analyses smartphone photos taken with a high‑quality magnifying lens and flags suspicious lesions for dermatologist review; the company says it can rule out melanoma with 99.9% accuracy. Deployed across the UK and conditionally authorised for three years, it has assessed more than 170,000 NHS patients since 2020 and is described by the maker as the first AI legally authorised to detect cancer without a doctor present. Proponents argue it can speed referrals and ease NHS pressure amid rising melanoma cases (≈17,000 diagnosed and ~2,300 deaths annually), while critics warn of data‑security, bias and empathy gaps. The episode highlights both the potential of AI to improve early detection and the need for robust regulation, transparency and clinical oversight as these tools move into routine care.