Health and life sciences data
Guides
Policy and lawTaking personal data out before a sale
AI buyers want what your data shows about the work, not about people. Here is what taking personal data out really means.
ImpactWhen company data saved lives
Three times ordinary company data, made for something else, ended up doing public good.
Stories
Fake patient notes that doctors can often spot
A public set of synthetic Portuguese clinical notes helps research without exposing real patients, but it is not perfect.
Data valueApple asks 200,000 people to grade a health chatbot
A study invites people to ask an AI their health questions, with a promise that the answers will not train a model.
Life sciencesHeart ultrasound images made from heart models
Researchers make moving heart scans from computer models, because real clinical sets are hard to build.
Life sciencesAuthors pull their own mouth-cancer data paper
A 37-author team withdrew a dataset paper after finding bugs in its script, saying the reported figures may mislead.
Policy and lawYou told a telehealth app it was private. Was it?
Five more US states join a federal case over health answers allegedly passed to advertising platforms.
Life sciencesBrain recordings from 7,272 people are rebuilt for training
A public set of brain-wave data keeps every owner's licence and still trains a model almost as well.
Life sciencesCardiologists invent patients so heart notes can be shared
Doctors wrote artificial hospital reports in seven languages, so AI can learn from clinical notes with no real patient behind them.
Life sciencesReaders could not tell AI themes from clinicians' themes
Three AI models read patient life stories and found themes the researchers missed.
AI and training dataMachine-written tests catch more AI failures than human ones
A method that grounds test questions in real-world evidence makes chatbots fail more often on medical advice and self-harm.
Life sciencesThin patient records learn to borrow from a map of medicine
A model decides, visit by visit, whether a patient record needs outside knowledge or can stand alone.
Data valueImage checks cannot prove where a medical scan came from
A test of synthetic chest X-rays at dataset intake says provenance has to be recorded, not guessed from pixels.
Life sciencesPublished papers hold more assay data than the databases
An AI workflow lifts results out of papers and builds a larger, versioned database for protein degradation drugs.
Life sciencesA tool that cleans health records for AI is put to the test
A new benchmark checks a chat agent that turns raw patient records into datasets ready for AI, and reports where it falls short.
Life sciencesA public trial registry trains a simulator that ignores the drug
A new test shows models built on registry data can score well without using the one field that matters.
Life sciencesA health-records AI model holds up in hospitals it was never trained on
Researchers pretrained a model on records from four health systems and tested it on two others, measuring what screening would really look like.
Data valueA rare public price for health data: what one patient record is worth
A filing with the US markets regulator puts a figure on de-identified patient records, and it is rarely this plain.
Life sciencesRegistry text alone carries some signal on which trials succeed
Language models read ten public fields per trial and beat the published benchmark, but not a trivial guess on every measure.
Sovereign and infrastructureUK science advisers ask government to name its important datasets
Advice to ministers on AI sovereignty puts data beside computing and skills, and asks for access to be tested in health and creative data.