Majid Afshar

@Majeans2011

Clinical informatics and ICU physician-scientist. Substance misuse and/or critical care research

Chicago
Vrijeme pridruživanja: siječanj 2011.
6 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @Majeans2011

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @Majeans2011

  1. prije 3 sata

    Now we can finally translate between our biostats and computer science colleagues

    The biggest difference between statistics and machine learning may be in language! So a few months ago I created this (inspired by ) but haven't made much progress since. Welcoming suggestions for improvements
    2 korisnika označavaju da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    9. pro 2019.

    On our message is clear: HEALTH IS A HUMAN RIGHT HEALTH IS A HUMAN RIGHT HEALTH IS A HUMAN RIGHT HEALTH IS A HUMAN RIGHT HEALTH IS A HUMAN RIGHT , stand up for !

    HEALTH IS A HUMAN RIGHT
    75 replies 2.270 proslijeđenih tweetova 3.644 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  3. 20. stu 2019.

    Packed house for presentation at . Great work.

    1 proslijeđeni tweet 5 korisnika označava da im se sviđa
    Poništi
  4. 17. stu 2019.

    All the credit to and his great Masters student Xin Su (who does not appear to be on Twitter).
    Poništi
  5. 17. stu 2019.

    Spending Sunday morning learning R on FHIR from the inventor of FHIR himself,

    6 korisnika označava da im se sviđa
    Poništi
  6. 17. stu 2019.

    Congrats and on making the top 4 in the n2c2 NLP challenge.

    1 reply 1 korisnik označava da mu se sviđa
    Poništi
  7. 9. stu 2019.

    The corporate business models of journals and publishers is obstructing access and progress in science

    Poništi
  8. proslijedio/la je Tweet
    9. stu 2019.

    10 tricks!

    5 replies 60 proslijeđenih tweetova 227 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  9. 5. stu 2019.

    Check out our paper that made it into the substance use PLOS collection:

    2 proslijeđena tweeta 1 korisnik označava da mu se sviđa
    Poništi
  10. proslijedio/la je Tweet
    1. stu 2019.

    Again, thanks to others for their input

    on the five myths of machine learning for clinical prediction models
    14 proslijeđenih tweetova 29 korisnika označava da im se sviđa
    Poništi
  11. proslijedio/la je Tweet
    30. lis 2019.

    A way to describe intelligence is that it is the power to produce abstraction. AI in the true sense would be Autonomous Abstraction. Current AI consists of recording abstractions generated by the human mind (via hard-coding rules, training ML models on human-labeled data, etc).

    30 replies 210 proslijeđenih tweetova 645 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    27. lis 2019.

    Develop new algorithms as a PhD student: $30k/year Use pre-built sklearn models as a data scientist: $120k/year Build regression models in excel as a hedge fund analyst: $200k/year Make pie charts as a CEO: $14 million/year

    251 reply 5.952 proslijeđena tweeta 27.570 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  13. 25. lis 2019.

    Every model for clinical decision making should assess for bias before deployment

    Poništi
  14. 25. lis 2019.

    Excellent paper that is relevant to clinical application for any risk prediction model

    NEW PAPER is out: Three myths about risk thresholds for prediction models (open access)
    1 proslijeđeni tweet 3 korisnika označavaju da im se sviđa
    Poništi
  15. 23. lis 2019.

    made simple and obvious

    9 korisnika označava da im se sviđa
    Poništi
  16. 17. lis 2019.

    We must do better for our teachers and children.

    4 korisnika označavaju da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    10. lis 2019.
    Odgovor korisnicima

    Provocative! Should we not refer to the slope of observed vs predicted as indicating 'calibration'? Agree with the notion that slope = 1 in itself is not sufficient to claim ‘adequate calibration’. A thread on perspectives on calibration of prediction models for 0/1 outcomes:

    1 reply 4 proslijeđena tweeta 18 korisnika označava da im se sviđa
    Poništi
  18. 16. lis 2019.

    Chicago public school teachers going on strike tomorrow for good reasons. Should I take my 6-yr old daughter to the picket line to see what her teachers are fighting for? Too much angst for her or the right teaching moment about advocacy...

    Poništi
  19. 16. lis 2019.

    Reviewer 2 in a recent submission tells me to trust readers will interpret subgroup analyses appropriately and share the results. O ye of too much faith. Maybe I should follow Richard peto’s lead and do a subgroup analyses based on horoscope signs ?

    Poništi
  20. 15. lis 2019.

    An excellent paper on how to take an algorithm into an algorithm-treatment study. Discrimination and calibration/recalibration is only the beginning. Then do decision curve analysis and, if acceptable, then consider designing a randomised study or quasi-experimental study

    Odgovor korisniku/ci @jd_wilko
    take a look at these ,
    3 korisnika označavaju da im se sviđa
    Poništi

@Majeans2011 još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·