Must read for anyone working with clinical/biomedical data and data visualization. Reveal, Don’t Conceal by @T_Weissgerber #dataviz
https://t.co/tNYX5AGQSM
El asunto del coronavirus está estos días todo el rato en la tele y los medios de comunicación.
Os propongo que demos un pasito hacia atrás y lo exploremos utilizando las 2 herramientas más poderosas que tenemos: la Ciencia y la Razón.
(HILO / 👇)
African grey parrots are smart enough to help a bird in need, and when they do it is adorable. Read the full story @newscientist https://t.co/ziSgQQzUQF (📹 Désirée Brucks/Auguste von Bayern)
A machine-learning model was able to distinguish healthy from cognitively impaired individuals roughly 80% of the time by how they used smartphone apps. https://t.co/jRMsHmufki
Our variable importance paper is out in JMLR!!
https://t.co/VNPI6uVVkL
Often, many prediction models achieve high accuracy. A variable could be important to some models, but not others. We study how important a variable could be to *any* model (in a pre-specified class).
(1/6)
Important large meta-analysis of individual patient data in @JAMANetworkOpen demonstrates that PFS is not a valid surrogate end point for OS in first-line treatment of advanced ovarian cancer, critical data to inform FDA approval decisions https://t.co/RqmXB5C91f
@daniela_oaks @raguirre_g Qué tal Daniela, es Cancer genomics, biomarkers, bioinformatics, en la escuela de Medicina del tec de Monterrey. Gracias por hacer la lista. Ojalá también podamos colaborar. Saludos!!
This feature by @mo_brouillette illustrates ongoing initiatives at hospital centers to train the next gen of physicians―"Doctors of the future will be expected to work closely with #ArtificialInteligence & therefore must understand #AI better" https://t.co/ZgAc4zlz1c @FAndreMD
Which cell types are found in tumors? What's the gene expression pattern of each? Want to find a fast way to develop hypotheses about genes involved in tumor-stroma interactions? Check out the lab's latest on a new method called TED, by @tinyichu.
https://t.co/KxCeTthb04
**new blog post ✍️**
overfitting: a guided tour 🚌
i present some intuition about overfitting, explain why it matters for both prediction and inference, and show why sample splitting is so important #statstwitter#DataScience
https://t.co/M7EllBnhUn
New personal project, @deepsleepml. As a long time sleep deprived dad of 5 kids, I wanna build a platform to share & implement resources for better sleep. I started sharing articles on sleep. plan to build an app. https://t.co/eBSrSAQvrY but it makes me sleep even less for now 🥱