Peng Qiu Wins CYTO 2017 Image Analysis Challenge

Contact

Walter Rich

Sidebar Content
No sidebar content submitted.
Summaries

Summary Sentence:

Competition held during the 32nd Congress of the International Society for Advancement of Cytometry

Full Summary:

No summary paragraph submitted.

Media
  • Peng Qiu, associate professor, recently won the conference-wide image analysis challenge held during the 32nd Congress of the International Society for Advancement of Cytometry. Peng Qiu, associate professor, recently won the conference-wide image analysis challenge held during the 32nd Congress of the International Society for Advancement of Cytometry.
    (image/jpeg)

Peng Qiu, associate professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory, recently won the conference-wide image analysis challenge held during the 32nd Congress of the International Society for Advancement of Cytometry held in Boston. At the time of the challenge closing, Qiu had the top solution for two of the four challenges making him the overall winner of the challenge. (http://www.proteinatlas.org/blog/2017-06-15/cyto2017-image-analysis-challenge).

 

In a time when vast amounts of bioimaging data are produced in labs around the globe every day, effectively extracting salient information from this growing resource of data is paramount to understanding complex biological questions. The CYTO 2017 Image analysis challenge proposed four tasks where the aim was to classify fluorescence microscopy images from the Human Protein Atlas database (www.proteinatlas.org) presented in a recently published study (Thul et al., Science, 2017). The images visualized immunostaining of human proteins and the aim of the challenge was to recognize the patterns of protein subcellular distribution to major organelles and fine substructures. (http://www.proteinatlas.org/learn/events).

 

A brief description of the dataset contained in each challenge:

Challenge 1:​ 1802 fields of view containing multi-label data for 2 protein localizations

Challenge 2:​ 20,000 fields of view containing multi-label data for 13 protein localizations

Challenge 3:​ 870 fields of view containing multi-label data for an additional 3 classes of localizations to be combined with the dataset from Challenge 2

Challenge 4:​ Using images in challenges 2 and 3, identify novel localizations that may be subtypes of the localizations in the previous challenges.

Additional Information

Groups

Wallace H. Coulter Dept. of Biomedical Engineering

Categories
Biotechnology, Health, Bioengineering, Genetics
Related Core Research Areas
No core research areas were selected.
Newsroom Topics
No newsroom topics were selected.
Keywords
BME
Status
  • Created By: Walter Rich
  • Workflow Status: Published
  • Created On: Jun 21, 2017 - 10:09am
  • Last Updated: Jun 21, 2017 - 1:07pm