SciFabric Logo
loading...

Hola! Nice to hear from your

Your message is on its way!

We’ll email back to you very soon.

Thanks for contact us :)

¡Hola! I'm Nick

Want to know how Scifabric can help you with your crowdsourcing project?

Get in touch

How crowdsourcing

can help beat cancer

Trailblazer is Cancer Research UK's PYBOSSA-powered crowdsourcing tool.

It has allowed thousands of people worldwide to help

scientists detect cancer cells with 90%

accuracy, after only 25 minutes of training.

Starting point

Trailblazer aimed to find out whether citizen science could be a

reliable method of accelerating cancer research

Citizen science holds true potential where researchers want to identify trends and patterns in large volumes of data.
This study investigates the efficacy of various training methods to improve accuracy in detecting several types of cancer cells.
Enhanced training and feedback interactive tutorials and easy-to-use interface were used to get people doing quality analysis, fast.

Why 's Technology?

PYBOSSA, our open source code

Flexible

With PYBOSSA you can fully customise your project: retain your brand, make your data private or public, add as many features as you need with our plugin infrastructure, import data automatically...

Dynamic, quick & easy

PYBOSSA provides ready-to-use templates, so you don’t have to start from scratch. It facilitates agile development and the release of new versions frequently. So you can go from prototyping to production in a flash.

Real time analysis

PYBOSSA allows real-time analysis. This helps analyse results as quickly as possible, saving you time, money and accelerating research.
Get in touch to know more

The project

Trailblazer at a glance

1000 volunteers

Over 1000 volunteers with no previous biomedical knowledge accessed Trailblazer, a web platform with no special software required.

25 minutes of training

After a 25 minutes training, volunteers had to analyse 10 sample images. Their scores showed they could distinguish between cancer cells and healthy cells in tissue samples with 95% accuracy. Try the Trailblazer demo here.

Evaluated by specialised pathologists

When it came to analysing real tumour samples, citizen scientists' scores were compared against consensus expert analysis to evaluate the accuracy.

And the results?

Trailblazer scored the highest accuracy reading of all Cancer Research UK’s projects based on crowdsourcing, proving that citizen scientists could be truly valuable to researchers.

Citizen scientists learned to detect clinically significant features of cancer cells. Their score was compared against consensus expert analysis to evaluate the accuracy. Volunteers' findings matched those of professional pathologists up to 90% of the time.
** Graphic shows example of Cancer Bladder analysis. Aggregating the analysis of 10 or more participants trained with both feedback and annotated examples, achieved accurate detection of cancer cells (Area Under Curve >0.90) for all samples tested. Full paper.
Trailblazer confirmed that citizen scientists could play an important role in accelerating research across a wide range of cancer types [...]. The levels of accuracy achieved have shown that citizen science can be a reliable, accurate and efficient method for biomedical research."

Andy Paterson

Digital Solutions Architect at Cancer Research UK and the technical lead on the Trailblazer project. Read the whole interview here

We are

We want to change the world for the better

Will you join us?

We like making people and technology closer. If you need to collect, analyse or enrich data, we can help you discover how crowdsourcing will accelerate your investigation and save you money.
Get in touch