Clinical work is fraught with decision making with high stakes and not merely enough evidence. AI has been touted as a panacea for all decision making as its ability to look at volumes of data and produce evidence can alleviate the need for clinicians in the first place. In this episode of our AI series, let's look at some of the use cases for AI in medicine:
- recommender systems that understand all medical evidence (in a field) and can look at an individual patient, find the most applicable evidence, apply it and make a recommendation
- "precision medicine" recommender systems that use machine learning to compare a patient with similar recent patients to choose the best outcome
- alerting systems that inform clinicians of critical events that pertain to their patients that they should not ignore, such as pathology/radiology reports and patient history
- pattern detection systems that identify important elements in images, text, and genetics
- visualisations and prioritisation of patient data
- evidence retrieval systems that find the most relevant evidence for a given patient
Many systems already exist on the market and are implemented. Adoption (and utilisation), however, are still very low. Is that a systemic problem that means AI is unhelpful in medicine? Stay tuned for the next episode.
Leave a Reply
Comments, questions or queries? Let us know!"