@NateSilver538 We are witnessing age related decline in performance. In the history of mankind no one has ever avoided that. President Biden is a really good man who is just too old to run again.
English
John Brush
280 posts
John Brush
@jeb1426
Practicing cardiologist, author
Virginia Beach, VA Katılım Ocak 2010
455 Takip Edilen394 Takipçiler
I’ve written in support of DANGER-Shock but @kaulcsmc is correct. The results should be confirmed. This decision is a big loss for EBM movement in cardiology.
Gregg W. Stone MD@GreggWStone
1/2 Following the DANGER SHOCK trial demonstrating reduced mortality with Impella in cardiogenic shock, the independent DSMB of the RECOVER IV trial has strongly recommended the trial be discontinued because of lack of equipoise. Therefore, we are ceasing enrollment in the trial.
English
John Brush retweetledi
Analysis of our regional real-world data showed rising hypertension prevalence and persistent racial disparities, providing a local call to action. Our analysis could be a model for other health systems. #AHAJournals @jeb1426 @yuan_lu1 @hmkyale ahajrnls.org/44LYLKV
English
@RichardLehman1 @dafraile @raj_mehta @stokemacca @sfinnikin What an important insight - "let time decide." In America, so much primary care has shifted to urgent care settings with poor follow-up that making a decision over time isn't an option. One has to either jump to a conclusion, or drop the opportunity to make a decision over time.
English
@dafraile @raj_mehta @stokemacca @sfinnikin Indeed! We will know that GenAI is mature for primary care when it advises there's no need as yet to put a label on a symptom cluster, but that it is safe to do nothing and let time decide. Like a good GP.
English
Perhaps the most intriguing GenAI paper so far for the future of general practice.
@dafraile @raj_mehta @stokemacca @sfinnikin
Joseph Kambeitz@joseph_kambeitz
Large-language model can outperform primary-care-physicians in a diagnostic interview. Notably: "Pediatric or psychiatry domains were excluded from this study..." Would this work with my patients? "Towards Conversational Diagnostic AI": arxiv.org/abs/2401.05654
English
John Brush retweetledi
"The value of a college education is not the learning of many facts but the training of the mind to think."
-- A. Einstein (1879 - 1955)
English
@quaesita I work with healthcare data where data privacy is a constant concern.
English
Hang on, what about data privacy?
Gagan Biyani 🏛@gaganbiyani
SQL is going to die at the hands of an AI. I’m serious. @mayowaoshin is already doing this. Takes your company’s data and ingests it into ChatGPT. Then, you can create a chatbot for the data and just ask it questions using natural language. This video demoes the output. 🤯
English
1/ A privilege to work with @hmkyale, @yuan_lu1,@RachelDreyerPhD, et al, looking at symptom phenotypes. Experts recognize diagnostic possibilities because they resemble memories called exemplars, which are mental representations of how features cluster as phenotypes.
English
4/ In STEMI pts with D2B time ≥90 minutes or no PCI, women had more symptom phenotypes than men overall and in the young patients, but not in the older patients. Initial recognition of diagnostic possibilities is important for D2B times and may be more difficult in women.
English
3/ Next we combined VIRGO and Silver AMI patients to look only at STEMI pts. Women had more symptom phenotypes, more symptoms on average and longer D2B times. Women had chest pain, but had more symptoms other than chest pain. Nausea was particularly prominent in young women.
English
2/ We examined the VIRGO and Silver AMI registries because these studies had details on symptoms in AMI patients. We examined how the symptoms clustered as symptom phenotypes and found that women, both young VIRGO and older in Silver, had more symptom phenotypes than men.
English
@hmkyale @YaleMed @YaleCardiology @yuan_lu1 @CircOutcomes @amjmed @AmJCardio 4/ In STEMI patients with either D2B time ≥90 minutes or no PCI, women had more symptom phenotypes than men in all pts and in the young patients, but not in the older patients. Initial recognition of dx possibilities is important for D2B times and may be more difficult in women.
English
@jeb1426 @YaleMed @YaleCardiology @yuan_lu1 @CircOutcomes @amjmed In a third study, we found 'more symptom phenotypes in young women were assoc w/prolonged D2B time or no PCI...greater frequency of nonchest pain symptoms in young women may have also slowed the recognition of STEMI and D2B times.' @AmJCardio sciencedirect.com/science/articl…
English
We have been doing a series of studies, led by @jeb1426, on sex differences in symptom complexity & phenotypes in acute myocardial infarction (AMI) and their impact on diagnosis & treatment. Some imp findings. #Cardiology #MedTwitter @YaleMed @YaleCardiology @yuan_lu1
English
@hmkyale @YaleMed @YaleCardiology @yuan_lu1 @CircOutcomes @amjmed @AmJCardio 3/ Next we looked at VIRGO and Silver AMI to look only at STEMI patients. Women had more symptom phenotypes, more symptoms on average and longer D2B times. Women had chest pain, but had more symptoms other than chest pain. Nausea was particularly prominent in young women.
English
@hmkyale @YaleMed @YaleCardiology @yuan_lu1 @CircOutcomes @amjmed @AmJCardio 2/ We examined the VIRGO and Silver AMI registries because they had detaileds on symptoms in patients with AMI. We examined how the symptoms clustered as symptom phenotypes and found that women, both young VIRGO and older in Silver, had more symptom phenotypes than men.
English
@hmkyale @YaleMed @YaleCardiology @yuan_lu1 @CircOutcomes @amjmed @AmJCardio 1/ A privilege to work with @hmkyale, @yuan_lu1,@RachelDreyerPhD, and others on these studies. Experts recognize diagnostic possibilities because they resemble tagged examples called exemplars, which are mental representations of how clinical features cluster as phenotypes.
English
@EiranGorodeski @HFSA Maybe a HTN/HF specialist: combining hypertension expertise with CHF expertise. Similar needs: timely follow up, attention to detail, and application of proven therapies.
English
