Combining citizen science and deep learning to amplify expertise in neuroimaging

The braindr web interface: braindr is hosted at https://braindr.us. Users may click pass or fail buttons, use arrow keys, or swipe on a touchscreen device to rate the image."

Researchers at the UW have developed Braindr, a web application for citizen scientists to inspect and annotate brain images by swiping left or right.

LEARN MORE

Nanoparticles could help us understand the relationship between structure and function in the brain

A team of UW researchers have recently begun to develop models that can integrate multiple types of data to better understand and predict structural and functional relationships in the brain.

LEARN MORE

Improving recreational opportunities with maps and tools from crowd-sourced and social media data

A selfie taken at Yosemite National Park. Photo credit: Zsolt Palatinus, Pexels

To help governments and community organizations meet the increasing and varying demand for outdoor recreation, a research team is innovating the use of crowd-sourced data and volunteered information from social media as instant and real-time data on recreation.

LEARN MORE

Creating reproducible and reusable biomedical models with model engineering

Researchers at the eScience Institute and their collaborators have received a grant to facilitate the development of technologies adapted from software engineering to improve the reproducibility and reuse of biomedical models.

LEARN MORE

Internet of the ocean

An octopus swims against a black background

UW scientists unravel the ocean’s mysteries with cloud computing, data science skills, and a sea of data.

LEARN MORE

How to discover asteroids with algorithms

Figure 1: Visualization of asteroids in the Solar System’s main Asteroid Belt. The circles represent the orbits of Mercury, Venus, Earth, Mars and Jupiter (see this animation on YouTube by Alex Parker).

Researchers have recently presented a new algorithm named THOR (for “Tracklet-less Heliocentric Orbit Recovery”) which employs coordinate transformations and clustering techniques to reduce the number of combinations needed to be tested to computationally tractable levels.

LEARN MORE

Community Snow Observations

Figure 1: Fiona Hill, Josie Hill, and Tessa Hill use a tape measure and ski pole to measure the snow depth in Mazama, Washington. Photo credit: David Hill

Citizen scientists collect snow data and submit it along with their location to a smartphone application; the data are then used to by scientists to check the accuracy of other regional snow datasets and model simulations.

LEARN MORE

nbdocker = Jupyter + Docker: simplifying reproducible research

nbdocker graphic

nbdocker is an extension to Jupyter that allows the embedding of Docker containers within notebooks. This allows the mix and match of different computing environments and programming languages.

LEARN MORE AND WATCH A VIDEO

Observing the cosmic dawn

HERA, under construction in 2018

We have developed one of the world’s most precise data analysis software suites to achieve the unprecedented spectral-spatial dynamic range needed to see the first stars and galaxies.

LEARN MORE

Fake profiles detection on online social networks

Graphic, Springer.com

A team of researchers, including Michael Fire, UW eScience Institute postdoctoral research fellow, has developed a machine learning solution to detect fake users on social networks. The research uses a generic unsupervised algorithm to improve the safety of these networks.

LEARN MORE AND WATCH A VIDEO

Shifts in marine microbial populations detected using statistical machine learning

Coverage of the SeaFlow data used in the change-point analysis. Black lines denote individual research cruise tracks and the heat-map shows the estimated quantity of chlorophyll at each spatial location based on satellite data.

The SeaFlow cytometer continuously profiles microbial populations across thousands of kilometers of the ocean surface during research cruises. The ‘multiple change-point detection’ method, developed by a UW team including eScience affiliates, can detect changes in scatter and fluorescence properties in these large datasets within seconds by using a dynamic programming algorithm.

LEARN MORE