From the organization:
Science Scribbler
Project avatar for Science Scribbler

Science Scribbler

Finished! Looks like this project is out of data at the moment!

See Results

Thank you for your contribution to our project! Our workflow on Desktop version is complete.
Please help us with Mobile app workflow. You can download the app from the Google Play Store or Apple App Store and you'll find Science Scribbler under the 'Medicine' section.

Results

Results from previous workflow of Science Scribbler project

We launched Science Scribbler: phase 1 last year. Over 1000 citizen scientists have volunteered to annotate the subcellular structure (organelles) inside a cell! With the contribution of citizen scientists, the annotation process that would take a researcher approximately a year was completed within a few months.

Images below show some data from the previous workflow. The left image of the top panel shows an image slice with annotations by volunteers. The right image of the top panel is a larger view of the left image (the green box). Each red blob corresponds to an annotation made by a volunteer to mark an organelle. The middle and bottom panels show some of the annotated data points in the 3D volume and an annimation showing a few image slices with data points respectively.


The next step of the project is to analyse the annotated data to answer whether there is any change in the subcellular structure (organelles) in Huntington's disease. The first step in doing this is to clean up the raw data. Annotation is a complicated and difficult task since there are different types of organelles with various intensities, sizes and shapes. In addition, differentiating the object (organelle) from the background (non-organelle) is particularly difficult in gray scale images due to a low signal-to-noise ratio. Because of this, there is variabilty and subjectivity in the original annotation data we collected. Before analysing this raw data, we would like to clean it up so that we can answer the biological questions more accurately and correctly.

This is where we need your help! In order to overcome the variability associated with the original annotations, we need your help to ensure a consensus emerges!

Read more about the previous workflow or classify here. You can also use the mobile app for the classifications.

Here are some of the previous academic papers we've published about segmentation (the process of marking and labelling areas within 3D volumes):