Thank you to @MarkKelly_News for the insightful interview on our recent study @VUMChealth.
With constant talk around #AI, it’s critical we test tools like these in the real-world and in clinical practice.
jamanetwork.com/journals/jam…
Developing risk models in healthcare is hard, implementing them effectively is harder!
We test two designs of a system prompting doctors to consider screening for suicide prevention driven by an automated model
/t
@vumcdbmi @VUMChealth@kbjohnsonmdtinyurl.com/3wtfcnf8
In clinical practice, there are lots of distractions (pagers, messages, phone calls). So interrupting doctors shouldn’t be done lightly.
In areas where interruptions wouldn’t be tolerated, less interruptive versions of tools like these might still make more sense.
/t
Much more to do including testing such systems at larger scale and refining them to prompt actions beyond screening in general.
These types of trials are critical to separate the hype around AI from their usefulness in practice!
@VUMChealth
/end
Dr Leigh Anne Tang, winner of the 2024 Brighter Ventures Award @VanderbiltU
She’s a newly minted PhD and rising star in healthcare and biomedicine
If you get a chance to work with her, do it!!
@laurentaudoly @vumcdbmi
medschool.vanderbilt.edu/bas…
Join us next week at the NIH Workshop on Ethics and Transparency in #AI for biomedical research.
We will be codeveloping best practices for synthetic data, multimodal data, foundation models, and more.
Free registration for videocast after the jump!
scgcorp.com/ethicalframework…
Can a novel #DeepLearning algorithm to predict #sepsis in the emergency department improve care?
Researchers from @UCSDHealth show proper implementation of the model can save lives of sepsis patients and improve quality metrics.
nature.com/articles/s41746-0…
Read our open access PsycheMERGE study to develop and validate scalable risk models to predict onset of bipolar disorder now out in @Transl_Psych
Three sites, three models, one paper!
@jorsmo @vumcdbmi @VUMCgenetics @GeisingerHealthtinyurl.com/36nhz76y
New study by @EHaroz and @jhucih in partnership with the White Mountain Apache Tribe
Prospective validation of a statistical risk model of suicide-related behavior within an associated care model.
A rare prognostic study in a critical domain
jamanetwork.com/journals/jam…
New study in @JMIR_PH led by Leigh Anne Tang with an incredible team at @TNDeptofHealth
Use of NLP to improve recognition of fatal overdose from autopsy reports. Implications for public health and the overdose crisis
@vumcdbmi
publichealth.jmir.org/2023/1…
Honored that our paper led by Jennifer Hong has won the Diana Forsythe Award at #AMIA2022. Forsythe was a groundbreaking anthropologist in informatics who wrote about women in tech, patient-centric informatics, and AI ethics long before they were cool. academic.oup.com/jamia/artic…