‘Minimalist machine learning’ algorithm analyzes complex microscopy and other images from very little data

Key tool for Chan-Zuckerberg-sponsored Human Cell Atlas project

These are images of a slice of mouse lymphblastoid cells; a. is the raw data, b is the corresponding manual segmentation and c is the output of an MS-D network with 100 layers. (credit: Data from A. Ekman and C. Larabell, National Center for X-ray Tomography.)

Neuroscientists devise scheme for mind-uploading centuries in the future

Representative electron micrograph of white matter region in cryopreserved pig brain (credit: Brain Preservation Foundation)

A high-density, stretchable, 32-electrode grid for neural recording and neurological disorder treatment

A potential Neuralink device? (see SXSW video)

Photo of a new soft, elastic, high-density 32-electrode grid for long-term, stable neural recording and treatment of neurological disorders. It’s based on a novel elastic material that's biocompatible and retains high electrical conductivity, even when stretched to double its original length. The 32 electrodes shown here are each 50 micrometers wide and located at a distance of 200 micrometers from each other. The fabrication procedure allows 32 electrodes to be placed onto a very small surface. The electrode grid is 3.2 millimeters wide and 80 micrometers thick. (credit: Thor Balkhed)

Super-resolution microscopy captures images in both space and time

High-speed “4D” views inside living cells

Cell image using color-coded depth

Metalens with artificial muscle simulates (and goes way beyond) human-eye and camera optical functions

Thin, flat structure promises to revolutionize eyeglasses, cameras, microscopes, and augmented and virtual-reality optics

A metalens (made of silicon) mounted on a transparent, stretchy polymer film, without any electrodes. The colorful iridescence is produced by the large number of nanostructures within the metalens. (credit:Harvard SEAS)

Measuring deep-brain neurons’ electrical signals at high speed with light instead of electrodes

“We will be able to watch a neural computation happen ... a step toward understanding what a thought or a feeling actually is.” --- Prof. Edward Boyden

Archon1 ft

Low-cost EEG can now be used to reconstruct images of what you see

Has promising uses for locked-in patients and forensics --- no expensive fMRI machine needed

(left) Test image. (right) Brain's image captured by EEG and decoded. (credit: Dan Nemrodov et al./eNeuro

Do our brains use the same kind of deep-learning algorithms used in AI?

Bridging the gap between neuroscience and AI

This is an illustration of a multi-compartment neural network model for deep learning. Left: Reconstruction of pyramidal neurons from mouse primary visual cortex. Right: Illustration of simplified pyramidal neuron models. (credit: CIFAR)

round-up | Two new wearable sensors may replace traditional medical diagnostic devices

Breakthrough technologies presented at AAAS annual meeting Feb. 17, 2018

throad sensor ft.

Neuroscientists reverse Alzheimer’s disease in mice

amyloid plaques ft

How to train a robot to do complex abstract thinking

Robot inspects cooler, ponders next step (credit: Intelligent Robot Lab / Brown University)

Are you a cyborg?

How to generate electricity from your body, bioprint a brain, and “resleeve your stack.”

Vertebral chip (credit: Netflix)

close and return to Home