George C. Linderman

@GCLinderman

MD/PhD student in Applied Mathematics . Interested in methods for denoising, analyzing, and visualizing high dimensional biomedical datasets.

Vrijeme pridruživanja: lipanj 2017.

Tweetovi

Blokirali ste korisnika/cu @GCLinderman

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

  1. proslijedio/la je Tweet
    28. sij

    Risk models can predict death or stroke after ACS, but do not perform well for *all* patients. Stultz Lab , , & developed “unreliability score” to determine if model’s results can be trusted for given patient cc

    , , i još njih 3
    Poništi
  2. proslijedio/la je Tweet
    23. sij

    The surgeon’s hand is visibly trembling. The scalpel’s blade glints as it catches the light. I’m a medical student, scrubbed in on the case. I’m not going into surgery, I know that. I’m here for him. He glances at me and notices I’m holding my breath. He grins. “Relax.” 1/

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    23. sij

    Google Dataset Search is now officially out of beta. "Dataset Search has indexed almost 25 million of these datasets, giving you a single place to search for datasets & find links to where the data is." Nice work, Natasha Noy and everyone else involved!

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    14. sij

    Check out the paediatric intensive care (PIC) database - 12,000 patients admitted from 5 ICUs in China - all publicly available!

    Poništi
  5. proslijedio/la je Tweet
    9. sij

    Great to see our heart failure GWAS paper published in Nature Communications today () - a huge effort from a brilliant team of collaborators providing insights into the genetic architecture of heart failure.

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    3. sij

    The predictive power of the microbiome exceeds that of genome-wide association studies in the discrimination of complex human disease

    Poništi
  7. proslijedio/la je Tweet

    Sir William Osler, the father of modern medicine, died on this day 100 years ago. In his honor, I'm posting some of his memorable quotes. First set on the practice of medicine

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    27. pro 2019.

    Earlier this month, I went to at NeurIPS. The talks were recorded, which is amazing. But I know you aren't going to watch 6 hours of video. Maybe you'll read 40 minutes of summary, though!

    Poništi
  9. proslijedio/la je Tweet
    26. pro 2019.

    Bayesian methods are *especially* compelling for deep neural networks. The key distinguishing property of a Bayesian approach is marginalization instead of optimization, not the prior, or Bayes rule. This difference will be greatest for underspecified models like DNNs. 1/18

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    26. pro 2019.

    This is a powerful example of how rigorously elucidating the molecular basis of a disease can change people’s lives and end suffering. Our scientific director Dan Kastner is a pioneer in the field of autoinflammatory diseases and his group continues to do breakthrough research.

    Poništi
  11. proslijedio/la je Tweet
    23. pro 2019.

    Interested in statistical models for scRNA-seq? Happy holidays! Two papers in : * GLM-PCA from and : * Our sctransform package (by Christoph Hafemeister) for normalization and variance stabilization:

    Poništi
  12. 20. pro 2019.

    Many thanks to Becht et al. for sharing their code and data. This is what open science is all about. .

    Prikaži ovu nit
    Poništi
  13. 20. pro 2019.

    UMAP is often said to be superior to t-SNE because it is better at preserving global distances. Dmitry and I showed that this is because by default, t-SNE initializes randomly whereas UMAP initializes with Laplacian eigenmaps.

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    19. pro 2019.

    Very excited to share our new preprint describing a method for ultra-high throughput single-cell RNA-seq

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    16. pro 2019.

    "The art of using t-SNE for single-cell transcriptomics" by and myself was published two weeks ago: . This is a thread about the initialisation, the learning rate, and the exaggeration in t-SNE. I'll use MNIST to illustrate. (1/16)

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet

    I'll start: - SR article about deep learning vs health care professionals: - This RCT baseline imbalance blog article: - This article about the big data paradox (technically 2018, but found it this year):

    Prikaži ovu nit
    Poništi
  17. 14. pro 2019.

    Are there any compelling examples of studies which controlled for a measured confounder (e.g. using regression), but because the model's assumptions were not met, the confounding was not controlled and an incorrect conclusion was reached?

    Poništi
  18. proslijedio/la je Tweet
    5. pro 2019.

    Thrilled to be a part of this project: "Single-cell connectomic analysis of adult mammalian lungs." Congrats and team (, , , et al.)--beautiful work!

    Poništi
  19. proslijedio/la je Tweet
    4. pro 2019.

    One of the first ideas I had the privilege to work on during my PhD is published! CHECK IT OUT: In a philosophical sense, think of this as a new, data-driven way to figure out how similar two observations are to each other!!

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    1. pro 2019.
    Odgovor korisnicima i sljedećem broju korisnika:

    One other important aspect, regardless of sample size, is the ethical advantage of increasingly randomizing more people to best scenario arms. This is, IMHO, a very compelling argument that should be more emphasized for clinicians 1/

    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:

    ·