. Massachusetts Institute of Technology (MIT) develop artificial intelligence which learns from patient data to optimise #chemotherapy and minimise adverse events, presented at #MLHC2018 news.mit.edu/2018/artificia… #tolerability #cancer #glioblastoma #machinelearning

BCS Prof. Michael Halassa chatted with Knowable Magazine about the role the thalamus plays in higher-level cognition. ow.ly/U0DC30md5lu McGovern Institute MIT Science

The secret to making ML systems less discriminatory: collecting better data. “Even the best mathematical models are no match for a biased dataset,” says MIT's Irene Chen. bit.ly/2S8yP5H (image v/@TechTarget)

#MGH researcher comments on new Weill Cornell Medicine study: "Researchers Develop New Strategy for Detecting Consciousness" the-scientist.com/news-opinion/r…

Correlation does not mean causation, but "if your thinking clearly separates association from causation, make sure your writing does too." Even if all you have is observational data. Miguel Hernán's paper entertained Emory Epidemiology journal club today ajph.aphapublications.org/doi/pdf/10.210…

2 years, 250 pages, 1,219 commits, and 78,480 words: I am very proud to say that today I published the 1st edition of "Interpretable Machine Learning". 🎉🎉🎉 Web: christophm.github.io/interpretable-… Leanpub: leanpub.com/interpretable-…

[Delayed cerebral ischemia after SAH] Kim et al.: "Epileptiform discharges increase after subarachnoid hemorrhage (SAH) during the high risk delayed cerebral ischemia time-frame." sciencedirect.com/science/articl… Yale Neurology MGH Neurology

Super excited to share this work combining multiple EEG features for DCI prediction after SAH. Special Thanks to WeiLong Zheng & Brandon Westover !

Very happy to share our recent work Nature Communications, where we develop a thalamocortical model using recurrent neural networks for context inference. nature.com/articles/s4146…