Differential (AI) Decisions
A new AI model uses electronic health records to make differential diagnoses
A new algorithm has been developed to help physicians make differential diagnoses (1). Current methods use Bayesian inference to determine the most likely diagnosis; however, Gerald Loeb, the engineer of the new tool, says, “What has been missing so far – and is provided by the algorithm – is a way to decide which clinical data to obtain at each point in the workup.”
He continues, “It prioritizes diagnostic options based on their likelihood to advance the process toward a definitive diagnosis vs. their cost in dollars, delay, and possible adverse events.” How? “It looks at the cumulative electronic health records (EHRs) of all patients in the database, including all tests that were run on those patients and their final diagnoses.”
Though the model assesses EHRs, pathologists have a vital role to play in the wider adoption of the model. Loeb believes it will “require substantial input from pathologists and laboratory medicine specialists to standardize the reporting of test results – particularly for newer modalities, such as genomic and antibody testing.”
- GE Loeb, J Biomed Inform, [Online ahead of print] (2021). PMID: 33711542.
During my undergraduate degree in psychology and Master’s in neuroimaging for clinical and cognitive neuroscience, I realized the tasks my classmates found tedious – writing essays, editing, proofreading – were the ones that gave me the greatest satisfaction. I quickly gathered that rambling on about science in the bar wasn’t exactly riveting for my non-scientist friends, so my thoughts turned to a career in science writing. At Texere, I get to craft science into stories, interact with international experts, and engage with readers who love science just as much as I do.
