Smadar Shilo

@smadarshilo

Pediatrician and Phd student at the Weizmann institute, segal lab

Vrijeme pridruživanja: studeni 2017.

Tweetovi

Blokirali ste korisnika/cu @smadarshilo

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

  1. proslijedio/la je Tweet

    AOP : Electronic health records to predict gestational mellitus (£)

    Poništi
  2. proslijedio/la je Tweet

    A new computer algorithm can predict in the early stages of pregnancy, or even before pregnancy has occurred, which women are at a high risk of gestational diabetes, according to a study in .

    Poništi
  3. proslijedio/la je Tweet
    23. sij

    Nice study from : a machine-learning approach to predict gestational diabetes mellitus (GDM) from 588,622 pregnancies with high accuracy (auROC = 0.85), plus a simpler model based on 9 questions that also is accurate and easy (auROC = 0.80).

    Poništi
  4. proslijedio/la je Tweet
    23. sij

    This work is an ideal demonstration of how Machine Learning should be used in Biomedicine. All steps are very well-thought-out. Do disease predictions from EHR, do not search for genes, they are hard to bring to the Clinics

    Poništi
  5. proslijedio/la je Tweet
    20. sij

    Superb work by the people at lab! Had so much impact even before being published here. Heavily used this framework with our recent study on prediction of gestational diabetes from nationwide EHRs:

    Poništi
  6. proslijedio/la je Tweet
    19. sij

    Description of large datasets and development of counterfactual prediction using complex models is the ongoing revolution. Herein a nice review discussing trade-offs between depth of phenotype and cohort dimension.

    Poništi
  7. proslijedio/la je Tweet

    Health data are being generated and collected at an unprecedented scale, but whether big data will truly revolutionize healthcare is still a matter of much debate, according to a Review in .

    Poništi
  8. proslijedio/la je Tweet
    18. sij

    Important paper. 2 key concepts: the axes of health data and the "deep cohorts". Deep cohorts should be considered by research agencies when it's about addressing questions that require multidimensional data (+ deep cohorts can be meta-analyzed and include a small control arm)

    Poništi
  9. proslijedio/la je Tweet
    13. sij

    Very proud to have been able to contribute to this amazing work, alongside Dr Becca Feldman of . Wonderful achievement led by . And now - working to implement it at .

    Poništi
  10. proslijedio/la je Tweet
    17. sij

    2) Paper led by Eran Segal group w/ our own Becca Feldman & top notch clinician team from Rabin MC. Already perfectly explained by Eran at: The best part in this, is we intend to go bench-to-bedside at Clalit, on both, asap! 3/3

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    16. sij

    • A detailed and informative read on big data in from et al in . IMHO, the importance of longitudinal data can not be understated— along with the need to increase the utilization of longitudinal analytics methods.

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet

    Leveraging nationwide electronic records from over 500k pregnancies in Israel, & colleagues develop a approach able to predict gestational with high accuracy at early stages of

    Poništi
  13. proslijedio/la je Tweet
    16. sij
    Odgovor korisniku/ci

    Thank you ! Here is the list of 11 tips for working with and the review

    Poništi
  14. proslijedio/la je Tweet
    16. sij

    RT : A fabulous paper for those interested in GDM and/of health data science. Routine health data and machine learning brings game-changing, early prediction of GDM. (Needs validation in multiethnic and higher prevalence populations though...)

    Poništi
  15. proslijedio/la je Tweet
    15. sij

    Great work on this review, led by and !

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    15. sij

    We discuss the great need for new approaches to drug development, and how human multi-omics data and physiological measurements at scale from deeply phenotyped cohorts may be one such direction, considered one of the most promising potentials of analyzing big data in medicine

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    15. sij

    We provide an overview of the geographical distribution of the main biobanks and cohort studies that are currently collecting and analyzing health data

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    15. sij

    We discuss how big data is analyzed, how massive datasets may achieve the potential of medical data analyses, how to bridge the gap between the data and our understanding and knowledge of human health We cover Descriptive analysis, Prediction analysis, Counterfactual prediction

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    15. sij

    We propose that medium-sized cohorts of hundreds or tens of thousands of people represent an interesting operating point, allowing collection of full molecular and phenotypic data on enough people We term these ‘deep cohorts’, as our own 10K project:

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    15. sij

    We discuss various tradeoffs in constructing human cohorts such as that between the scale of the data gathered (axis N) and the depth of the data (axis D), and show where different cohorts lie, including mega biobanks (e.g., UKBiobank), genetic datasets, EHR datasets, etc.

    Prikaži ovu nit
    Poništi

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:

    ·