More data is being generated than can be analysed - but enough of it is being studied to spot patterns in behaviour and genes to speed up drug discovery and development.
Though the data scientists need to make sure they feed in the right data. For example, in a study designed to detect hip fractures, it turned out the AI was using the patient's age and particularly the imaging device model to make its prediction. This worked because portable scanners are only used then the patient is too immobile to get to a hospital - and this limitation turned out to be a bigger indicator than any details on the image itself.
Evidence Based Medicine - Wikipedia
Amazingly it wasn't until the 1990s that research results became an important part of decision making, rather than an individual doctor's clinical judgement.
Lung cancer diagnosed by Artificial intelligenceBBC - 20-May-2019
AI and doctors working together can be more effectiveRead more...
IBM stopped the sales of Drug-Discovering Watson AIFuturism - 19-Apr-2019
Powerful technology in the healthcare field must improve patients’ healthRead more...
FDA proposes new rules for locked and adaptive AI algorithmsThe Register - 03-Apr-2019
Regulators will struggle to keep up with improvements in AI diagnosisRead more...